Hi I stopped normalizing my vox years ago because it's destructive but I'm just not getting a sound I like. I have high-end ribbon mics & use an in-line pre but still have to get so close I get the proximity effect. I'm thinking of recording further away for a more natural sound & normalizing for a better level. The only drawback i hear in a test is some low-end noise I think i can notch out. Is there any reason why i shouldn't be doing this? I don't want to get new equipment at this stage of life. Thanks for any thoughts!
normalize vox?
- Thread starter paresh
- Start date
Ujn Hunter
Active member
Normalizing? Raising the Levels? Are you using EQ and some Compression? Those things will probably do more for your Vocals than "Normalizing". Also... why are you choosing Ribbon mics over Condenser or Dynamic mics for Vocals?
rob aylestone
Moderator
Just buy a preamp, then you can record further away? Normalising just changes the maximum level, the ratio of signal to noise stays the same - so the volume goes up and so does the noise.
Why not ribbons? You saw a lot of 4038s in BBC talk studios and the Beyer twin ribbon is very well thought of.
Reslo RBs were good enough for the Beatles in the cavern!
Dave
TalismanRich
Well-known member
I'm wondering about getting a "high end ribbon mic" and wondering what the rest of the recording chain is. A good ribbon into inline pre into an interface with 100+dB noise level at 24bits should be absolutely workable without being close enough for the proximity effect to be severe.
It is actually a bit of a myth that ribbons have a very low sensitivity. The Golden Age AP R1 Mkll has a sensitivity of -52dB and that is 2.5dB hotter than an SM58! Most ribbons have a similar sensitivity except the venerable Coles 4038.
No, the output is an issue because we naturally use ribbons at 'capacitor' distances when they are at a 10 even 20dB disadvantage. Close up with a really effective pop shield most ribbons will be no more trouble than a dynamic. The bass boost can easily be corrected post tracking. Have to do that with directional dynamics quite often anyway.
Re "normalizing"? I only know it in Samplitude where the degree to which you approach 0dBFS can be set and saved. Often however, if I just want a few dBs boost I just crank the relevant fader(s) and "save as". And "destructive"? Pretty sure I can undo normalization in Sam?
Dave.
No, the output is an issue because we naturally use ribbons at 'capacitor' distances when they are at a 10 even 20dB disadvantage. Close up with a really effective pop shield most ribbons will be no more trouble than a dynamic. The bass boost can easily be corrected post tracking. Have to do that with directional dynamics quite often anyway.
Re "normalizing"? I only know it in Samplitude where the degree to which you approach 0dBFS can be set and saved. Often however, if I just want a few dBs boost I just crank the relevant fader(s) and "save as". And "destructive"? Pretty sure I can undo normalization in Sam?
Dave.
Thanks for the comments! I can try working w compression more. I don't think EQ will help levels. Ribbons are my only mics, it's just a small project studio. The pres are Titan FET heads but I don't think they help much. Maybe I'm expecting too much? I've priced other preamps but they are expensive. i could trade one of the ribbons for a dynamic mic but I'm afraid I'd miss the natural transparent sound. I think condensers might be too bright or too much sibilance for me. Does a pop filter/wind screen help the proximity effect?
Random dude
Well-known member
Since it wasn't asked or mentioned, I guess I will.
What do you consider a "good level"? What makes you think your level is too low? When you're tracking and mixing, the level almost doesn't matter as long as it doesn't clip and as long as it's not so low that the noise floor is louder. With digital recording, you would have to be REALLY low for that to be an issue. If the other instruments are too loud for the vocals, just turn them down.
Volume has nothing to do with sound quality. It might SEEM that getting a "good signal" makes thigs sound better, but that's just psycho-acoustics.
Once you've mixed your song, the next step is where you adjust overall volume. It's one of the things that happens in the mastering stage. Until that point, don't worry about levels being too low. Just turn up your monitors/headphones, etc...
What do you consider a "good level"? What makes you think your level is too low? When you're tracking and mixing, the level almost doesn't matter as long as it doesn't clip and as long as it's not so low that the noise floor is louder. With digital recording, you would have to be REALLY low for that to be an issue. If the other instruments are too loud for the vocals, just turn them down.
Volume has nothing to do with sound quality. It might SEEM that getting a "good signal" makes thigs sound better, but that's just psycho-acoustics.
Once you've mixed your song, the next step is where you adjust overall volume. It's one of the things that happens in the mastering stage. Until that point, don't worry about levels being too low. Just turn up your monitors/headphones, etc...
Let's get back to basics Paresh? What interface are you using? You say the Fetheads don't help much? In what way? My son uses one with an SM57 on classical guitar and a Behringer UMC 204HD* and gets good results with the mic 400mm from guitar.
Why do you need to be so close? Lack of level or poor room acoustics or ambient noises? If a level problem, what are you aiming for? The received wisdom is to record at an average of -20dBFS. If a bad room sound more duvets! If "noises off" then little can be done unless you can record when there is no noise. Son used to do that here at 2 and 3am.
If what you want is a 'one shot' recording, 'in the can' at -6dBFS then you will struggle IMHO. Now, I know I shall get shot at for this but a small mixer could give you EQ even high pass filter on the way in. Most modern mixers of the Behrry stamp have pretty good pre amps and should be fine mated with the Fethead. You might even find a mixer with EQ, HPF and a compressor.
*really quite good pres but not good enough unless the 57 was so close he clouted it!
"Hey Dude!" you are faster on the qwerty than this old duffer!
Dave.
Focusrite scarlett i18 8. I thought I needed to be so close because otherwise the waveform is so low but re the previous post i will rethink that. I thought the pres weren't helping much because of the low waveform level. I'll check the level & see what I am recording at. Thanks.
Random dude
Well-known member
It's ok. You always give great advice and insight...for an old duffer, of course.
Random dude
Well-known member
Not trying to give you a hard time. We're just discussing and trying to find a solution. So, let me just ask...When you say "it seems too low", what do you mean? Too low compared to what? Songs on Youtube? MP3's of pro artists? It's totally normal that your tracks will be lower than those things until you finish your mix and master it. Like I said before, don't worry about volume at this stage of the process.
Also, why do you have all the faders at unity? The definition of "Mixing" is.......well, "Mixing". Some faders will be lower than others. You can have some faders up at unity (even though I never do, but that's besides the point), and other faders down at -10, -16, or even -20. Unless I misunderstood and you were just using that as an example.
It will only increase the noise floor by the same amount that it will increase your mic volume. If your noise is 50db's lower than your signal, it will still be 50db's lower if you increase the trim. The proportion stays the same.
rob aylestone
Moderator
The thing I have noticed is that the ballistics of digital meters seem to vary so much, so our levels are always subjective. I've discovered that recording with a decent preamp, you can have very low levels that are totally fine when normalised up, and adding gain at any stage seems to work pretty well - everything in our chain - the preamp knobs, the DAW input fader, the channel faders and the masters seem to allow huge scope in levels. This was so different in analgue days when you had to optimise every stage between in and out to maximise signal and minimise noise.
Random dude
Well-known member
Yes, with "tape noise" not being an issue, noise is almost a non-factor. Unfortunately, because of left over beliefs from the days of tape/analogue, you still have people that think you need to record "As close to 0db as possible without going over." Personally, I almost never even look at the level when I lay down a track. I know I record low enough that I never get anywhere near "0db" so I don't worry about it at all.
Farview
Well-known member
It kind of sounds to me like you are confusing level with volume.
You need to record a certain level to get the best signal to noise ratio, then adjust the volume in the mix to make everything listenable. It is two separate processes.
The Focusrite pre amp section has a gain spec' of 56dB and that is a bit low for a dynamic or ribbon mic at a distance for vocals (unless you are Pavarotti re born!) I am also assuming that if you max out the gain the pre amp hiss is audible?
The idea of a 'pre pre amp' is that you back off the AI gain until there is no noise and the extra pre more than makes up the gain. You have a total of some 80dB on tap so with some careful setting you should be able to get a good level with the mic beyond the proximity effect distance. Whether other factors, room noise and acoustics then intrude is another matter.
Dave.
The idea of a 'pre pre amp' is that you back off the AI gain until there is no noise and the extra pre more than makes up the gain. You have a total of some 80dB on tap so with some careful setting you should be able to get a good level with the mic beyond the proximity effect distance. Whether other factors, room noise and acoustics then intrude is another matter.
Dave.
TalismanRich
Well-known member
Are you hearing electronic noise or ambient noise? Unless I record at a ridiculously low level, my ambient noise level is far higher than any preamp/mic noise, even using dynamic mics like my SM57 or my Senn e935. I'm going into a Tascam 16x08 which is probably similar in spec level to your 18i8. I don't even have a Cloudlifter/FETHead. I've done vocals with the e935 just plugged into the Tascam and it was perfectly usable singing around 4-5 inches away.
Not seeing a big waveform on the screen doesn't mean that you don't have 70 or 80dB of Signal to System Noise. Most of the times, if I normalize a signal so that I can see the actual waveform in Reaper without using the zoom function, it's only 15-20dB. It's handy to see the waveform if you're editing so that you can cut and paste with a bit more precision, but that's the main reason I do it, unless I get a track that is unusually low, and I have trouble getting enough gain with the volume maxed out compared to other tracks.
In any case, doing a normalization should not affect the sound at all. Remember the saying that a rising tide raises all boats. The differential of loud to soft is still the same.
If you are concerned about the base noise level, then do a recording with the mic set up with the "low" waveform, then be quiet and see what the actual noise floor is (really easy to see with something like the Blue Cat Analyzer). In many cases, you're looking at something -50 to -60dB in most rooms.
Not seeing a big waveform on the screen doesn't mean that you don't have 70 or 80dB of Signal to System Noise. Most of the times, if I normalize a signal so that I can see the actual waveform in Reaper without using the zoom function, it's only 15-20dB. It's handy to see the waveform if you're editing so that you can cut and paste with a bit more precision, but that's the main reason I do it, unless I get a track that is unusually low, and I have trouble getting enough gain with the volume maxed out compared to other tracks.
In any case, doing a normalization should not affect the sound at all. Remember the saying that a rising tide raises all boats. The differential of loud to soft is still the same.
If you are concerned about the base noise level, then do a recording with the mic set up with the "low" waveform, then be quiet and see what the actual noise floor is (really easy to see with something like the Blue Cat Analyzer). In many cases, you're looking at something -50 to -60dB in most rooms.
Random dude
Well-known member
Oooh....I like that. Never heard it before. But now it's going to be one of MY lyrics.
Attached are two recordings of speech (I am NOT going to sing!) using a Relso RB. The AI is a MOTU M4 and the mic gain is at max.
You can see that the primary recording is at a very low level, barely peaks to neg 30. The normalized version peaks to about -6dBFS and there is a 'silent' part at the end where you can judge the S/N ratio. For a scratch setup and a 70 year old mic, not too foul?
Note, the mic is 30-50 Ohms but goes through a Sowter 8754 transformer with a 1:4.5 step up ratio giving around 12dB of 'noise free' gain. You can hear some cyclic LF noise and I would guess that is the laptop's fan. There will also be some 50Hz in there.
Hope that gives you some sort of 'benchmark'?
Dave.
You can see that the primary recording is at a very low level, barely peaks to neg 30. The normalized version peaks to about -6dBFS and there is a 'silent' part at the end where you can judge the S/N ratio. For a scratch setup and a 70 year old mic, not too foul?
Note, the mic is 30-50 Ohms but goes through a Sowter 8754 transformer with a 1:4.5 step up ratio giving around 12dB of 'noise free' gain. You can hear some cyclic LF noise and I would guess that is the laptop's fan. There will also be some 50Hz in there.
Hope that gives you some sort of 'benchmark'?
Dave.