What things have you learned recently?

[Gabe]

1. I've learned more about proper gain structure, to try and record to digital at around -18dBFS. I already understood this to a certain extent, but I would be tempted and be recording closer to -10dBFS or maybe even a little higher. It's amazing how much this simple thing has opened up my recordings.

Just a quick question about this method: why is it better to not hit the converters as hot as possible when recording in digital? obviously clipping will wreak havoc on the quality of your signal (squaring off all waves over 0db mark and adding distortion) but when recording at 24bit if only half the gain is used its basically as if you are sampling at 12bit correct? and if you are using less than a quarter you are down to 6bit (less than telephone??).

how does recording a quieter signal open up a recording?

[drumsound]

1. I've learned more about proper gain structure, to try and record to digital at around -18dBFS. I already understood this to a certain extent, but I would be tempted and be recording closer to -10dBFS or maybe even a little higher. It's amazing how much this simple thing has opened up my recordings.

Just a quick question about this method: why is it better to not hit the converters as hot as possible when recording in digital? obviously clipping will wreak havoc on the quality of your signal (squaring off all waves over 0db mark and adding distortion) but when recording at 24bit if only half the gain is used its basically as if you are sampling at 12bit correct? and if you are using less than a quarter you are down to 6bit (less than telephone??).

how does recording a quieter signal open up a recording?

By not adding excess harmonic distortion.

[Gabe]

how does recording a quieter signal open up a recording?

By not adding excess harmonic distortion.

Hey tony thanks for the quick answer... i'm still kind of confused though. So even before the channel clips, there are still amounts of inherent harmonic distortion that are brought out when hot signals are attempting to be converted?

does this mean we use 24bit so when can get 16bit quality with out the distortion a 16bit system would yield when attempting to record that hot?

thanks for being patient guys... this seems like a big concept i want to get it straight.

[drumsound]

Everything can distort and just because you don't see a red light and because it's not ripping your head off doesn't mean that there isn't distortion. Part of the reason to stay at 0vu goes along with the John Scrip quote above.

Read this and part 2. Even -18DBFS in 24-bit digital is better resolution that 16-bit at 0DBFS.

Just to stir the pot a bit, sometimes distortion that isn't obvious (subtle harmonic distortion) can benefit a recording.

Roy Thomas Baker used harmonic distortion and generally pushed levels to make brilliant recording for instance.

[Gabe]

Read this and part 2. Even -18DBFS in 24-bit digital is better resolution that 16-bit at 0DBFS.

Hey Tony thanks a lot for clearing that up, i really appreciate it. This is the most recent thing i've learned about recording!

[chovie d]

What things have you learned recently?

1. follow your own dreams
2. dont depend on anyone else for anything
3. if you are healthy be thankful, dont take it for granted and enjoy each day

[Kasey]

24 bit audio (and even 20 bit) is to the point where it really isn't necessary to "use all the bits".

the "opening up" occurs due to the micro-transients that aren't registered on the average meters having headroom to exist within. at least that's my understanding. makes sense. and the difference in sound is there so i cant argue..

[paranoidandroid]

Gain is a logarithmic function, every bit equals 6db equals double the loudness of before. Thus if you made a recording half as loud as it was before you would only be losing one bit of resolution, not 12. There is also very little chance that the room you're recording in is actually quiet enough to capture even 90db of dynamic range, and even if it was, the microphone/ad converter couldn't pick up more than 120 or so. That leaves a good 24db of headroom that I have found is best used just like it is in analog land, leave some space for those questionable plugins to do their magic.

Thing I keep coming back to recently: classic high end gear will give you a sound that is standard and acceptable, it makes your job easier but at the same time, because it is classic, restricts the possibility of 'original' sounds. I think investing in a monitoring/treatment setup that really really really allows you to understand what's going on allows for the greatest amount of confident creativity, if you trust your ears you can use as many tools as you need to get the sound you want instead of being happy with the sound you're given through 'classic' gear. Frankly I question why I would want a pair of distressors when my speakers cost half as much, shouldn't they cost the most? especially if they're active?

[Kasey]

now now, back to the original question.

[JASIII]

I've learned that just a small amount of bass and guitar go a long way when adding them to a parallel compression bus.

I've finally accepted that putting a shallow chorus on panned guitars does sound good and beef up the sound (I'm normally anti-chorus, I just don't like it).

[mwingerski]

I've learned that spending too much time on the computer reading about things takes away from time I can spend creating music and actually working (but i still do it anyway)

I've learned that all the gear in the world isn't as good as a well played part on a nice sounding instrument. A great performance makes a great mix.

I've learned it's far more important for an artist to feel comfortable and play well than it is to get separation and sonic clarity.

I've learned that great gear makes good stuff sound much better.

I've learned that there is not a strong correlation between the quality of music or recording and its sales impact in the marketplace and I've also learned that it's not worth getting angry about this.

[markmeat]

I've learned that there is not a strong correlation between the quality of music or recording and its sales impact in the marketplace and I've also learned that it's not worth getting angry about this.

Amen.

[akg414]

1. I've learned more about proper gain structure, to try and record to digital at around -18dBFS. I already understood this to a certain extent, but I would be tempted and be recording closer to -10dBFS or maybe even a little higher. It's amazing how much this simple thing has opened up my recordings.

+1. amazing how much it really does help.
-ryan

Can you ellaborate on this a bit? Keeping tracking levels at -18...

Thanks.

[Kasey]

1. I've learned more about proper gain structure, to try and record to digital at around -18dBFS. I already understood this to a certain extent, but I would be tempted and be recording closer to -10dBFS or maybe even a little higher. It's amazing how much this simple thing has opened up my recordings.

+1. amazing how much it really does help.
-ryan

Can you ellaborate on this a bit? Keeping tracking levels at -18...

Thanks.

You should check out those threads I mentioned. They'll explain it much better than I can.

Basically, when you record you want to keep your signal around 0dBVU. That is were it generally sounds the best no matter what you're recording. However, in digital we're talking dBFS.. not dBVU.
0dBFS does not equal 0dBVU. It depends on the calibration of your digital unit... but usually its about -18dBFS = 0dBVU

Someone along the line forgot this and people generally assume that you should get as close to 0dBFS as you can ("that's what they used to do" ...except they weren't using digital... and they were metering in dBVU, which we generally are not). This trend began when digital sucked and there really was a reason to "use all the bits" but now it doesn't matter with digital quality so high. If we continue to try and "use all the bits" we're just destroying headroom for micro-transients that your typical peak meters cannot detect, which is what gives our sound an "open" quality.

It's OK to go over -18dBFS... during the days of tape they would go over 0dBVU all the time, but not by 18dB like we are doing when we try to get as hot as we can without clipping.
-13dBFS is the equivalent of +5dBVU, which is fine for peaks. But if we're going all the way up to -1dBFS, that's equivalent of +17dBVU! Bad idea.
You should try to keep the meat of your signal around -18dBFS, though peaks can go a little above with no damage done.

There are of course exceptions. Some people will tell you that tracking snare and kick drum hotter than other things gives their sound a more punchy quality. (for these things, we're not necessarily going for "open" as much as that punch.. so we can cross the line a little bit). And also, drums have always been recorded hot, regardless of medium... everyone just wants hot sounding drums.

Your digital gear may not be calibrated to -18dBFS... that's just the norm. So read up and find out. It will really help your recordings, I promise.

Cliff Notes
-Things sound best at 0dBVU
-Digital doesn't measure in dBVU, it measures dBFS
-Usually, -18dBFS=0dBVU
-Thus, in digital, -18dBFS sounds best in most cases
-This is due to the headroom given to micro-transients, which is what makes the sound "open up"



peace.

[adam]

Yeah, great thread - what a great way to learn. Not recent revelations, but some significant ones...

- Mixing at low volume is a good thing
- Patchbays are built by angels
- My decision to switch to Pro Tools was a good one (for me)
- Good cable makes a difference
- Delay is more useful than reverb
- Silvertone amps are incredible to record
- Don't sweat sample rates and bit depths
- A well place Radioshack mic is better than any poorly placed mic
- Don't ever sell anything unless its just shit

Adam