Tracking - Levels - Page 4 - Gearslutz.com

Gearslutz.com

All Advertisers
Go Back   Gearslutz.com > The Forums > Music computers


Tracking - Levels

New Reply New Reply Thread Tools Search this Thread
Old 24th July 2008   #91
Gear maniac
 
Joined: May 2008
Location: DC/Balt
Posts: 167

Quote:
Originally Posted by Pfhuck View Post
So, we've already established that 24bits of resolution is more than adequate....

Isn't keeping peak levels down (around -10ish) worth the peace of mind??

Who wants to stress about digital overs while tracking?

"Oh sorry dude, you got waaaay too emotional on that last take and my a/d clipped. Gotta do it over again. SORRY!"

But then again, maybe you want that sound??

I do sometimes.

Laters,
Pfhuck
amen to that! nothing is worse than catching a great performance only to have the levels clip at the pre or converter resulting in an unusable sound. this makes me feel very useless and i try to make it NOT happen as much as possible. After reading this thread i am def going to track lower levels on 24bit sessions. i have always understood the signal to noise thing but still i think i'm tracking a little too hot in general. werd up! good thread.

p.s. normalizing is totally lame!!! save it for the mastering engineer!

E
EvilE is offline   Reply With Quote
Old 25th July 2008   #92
Gear Head
 
Joined: Sep 2007
Posts: 69

Very interesting read.
I have always used the -6 as 0db method mentioned earlier.
If i were to record (into logic) peaking at -18dbfs, I think it would lead to problems with plugins. For example, if I wanted to limit a snare with sonnox limiter, I would have to push the input gain to +22 if I wanted about 4db of limiting then pull the output back by 18-20db to get back to about -18 peak. For eqs, delays, reverbs etc the level makes no difference but for dynamic processors there would be a lot of pushing the input gain to drive the effect then pulling back the output. Maybe i am missing the point.
I always use prefade metering to avoid clipping effects and make sure the master peaks at about 6 so, taking these points into account, would there be any advantage recording at -18dbfs.
Michael
mickrich is offline   Reply With Quote
Old 25th July 2008   #93
Gear nut
 
WinterSun's Avatar
 
Joined: Jun 2008
Location: Alton, IL / St. Louis
Posts: 137

Noise Floor!

I've only seen like 1 person talking about noise floor. I was always taught to track with high levels to keep the noise floor down. If you track too low and have to raise the level, the noise floor rises with it.

Does no one else find this to be an issue??
__________________
Winter Sun Recording

Steve Cox
Owner/Engineer
WinterSun is offline   Reply With Quote
Old 25th July 2008   #94
Lives for gear
 
synthoid's Avatar
 
Joined: May 2006
Posts: 2,795

There are two quite distinct issues here, getting muddled together.

The first is this: what's a good peak level to aim for at the A/D converter when tracking?

The second is this: should you aim at a particular peak level at all, or should you rather (as the OP suggested) capture each track at a volume that gives you a rough mix with faders at unity gain?

I don't want to comment on the first one, haha, because it's such a slugfest. I suggest you listen closely to your A/Ds, analog front end, etc., and make a decision based on the gear you have at hand. It's a complicated equation because many ADCs let you set or select the reference level (the analog level that corresponds to 0dbFS).

The second one, however, strikes me as questionable recording practice. You can indeed run into noise problems if you have to boost the level of a part that you tracked at too low a level. Someone said that pulling faders down hard in PT HD sounds bad!? Yikes. That's unbelievable to me.

Tracking is tracking, it has its own rules of good practice. Mixing is mixing, it happens later. Rough mixes and notes are the right way to communicate from tracking engineer to mix engineer IMO.

-synthoid
synthoid is offline   Reply With Quote
Old 25th July 2008   #95
Lives for gear
 
Joined: Dec 2002
Location: U.K
Posts: 2,006

Quote:
Originally Posted by mickrich View Post
Very interesting read.
I have always used the -6 as 0db method mentioned earlier.
If i were to record (into logic) peaking at -18dbfs, I think it would lead to problems with plugins. For example, if I wanted to limit a snare with sonnox limiter, I would have to push the input gain to +22 if I wanted about 4db of limiting then pull the output back by 18-20db to get back to about -18 peak. For eqs, delays, reverbs etc the level makes no difference but for dynamic processors there would be a lot of pushing the input gain to drive the effect then pulling back the output. Maybe i am missing the point.
I always use prefade metering to avoid clipping effects and make sure the master peaks at about 6 so, taking these points into account, would there be any advantage recording at -18dbfs.
Michael
Yes this is a very good point and a limitation of the workstation environment (among many that we had tied down in consoles decades ago).

The point is that all signal ultimately has a relationship to the real world level - which is in the end what we will hear. Tracking at lower levels to give yourself back the headroom that the W/S designers overlooked is fine - except that plugs that need to know 'real values', in the absence of a real operating level variable, are necessarily still calibrated to 0dBFS - the level of the meter 'red lights'.

This means that stuff like the OX Limiter that requires significant gain to get certain stylised sounds may not have enough gain to make up for your headroom level loss - so you may lose the ability to achieve certain highly compressed sounds all in one go etc..

In consoles like the Sony OXF-R3, we therefore had a 'headroom global value' that was set on boot up options, which compensated for the gains in all signal paths - and passed 'real operating level' values to all internal processing, so that they 'knew' the intended output target level and behaved appropriately, whatever headroom was set up. This is of course how things should be done - and this is but one of many subtleties that have been sadly lost in the changeover to workstations.

It is stuff like this that happens under the hood (and never reported in brute specifications) that truly separated professional apps from consumer stuff. Basically it's that 'it just always works' experience you get from stuff truly designed for professional use. It's purpose was to relieve you of the nitty gritty and frustration of bad behaviours and time consuming measures spent avoiding them - and let you get on with producing art :-)

The idea used to be that you left the technical detail to those that had the experience, you got on with the job - and was not required to spend ages batting around trying to unearth hidden issues, in a desperate attempt to acquire the experience and knowledge that had been left out of your system :-(

And of course - all that kind of stuff was what largely made professional kit so much more expensive....
__________________
Paul Frindle www.proaudiodsp.com
Paul Frindle is offline   Reply With Quote
Old 25th July 2008   #96
Lives for gear
 
peeder's Avatar
 
Joined: Aug 2006
Location: No longer participating here.
Posts: 6,705

You should listen back carefully post-converter to assess proper gain staging for your particular chain. You should not fall to "meteritis" where you are a slave to some moronic rule of thumb foisted on you by some disembodied phantom pecking at a keyboard thousands of miles away (and btw I'm not pointing at anyone in particular). Do not sacrifice the huge investment you've made in a quality signal path trying to hit someone else's ideal meter reading.

Use analog attenuators only when the desired sound dictates doing so. Clipping a converter may sound good on some signals. So might noise. Free yourself from the tyranny.

Once ITB, gain and trim can be applied at any stage with virtually no consequence (at least in floating point systems). Many plugins have saturation models just like analog processors do, and each plugin will have its own levels and behaviors. Most such plugins have an input trim knob, and if not, there is a trim/gain plugin. Do not sacrifice this behavior of plugins either; some idiot may overdrive a plugin obliviously, and run to their computer and peck in, "ALL PLUGINS SOUND AWFUL WHEN OVERDRIVEN." Have a laugh.
peeder is offline   Reply With Quote
Old 25th July 2008   #97
Lives for gear
 
Joined: Jul 2006
Posts: 1,131

Quote:
Originally Posted by WinterSun View Post
I've only seen like 1 person talking about noise floor. I was always taught to track with high levels to keep the noise floor down. If you track too low and have to raise the level, the noise floor rises with it.

Does no one else find this to be an issue??
Not really. Your dealing with 144db of dynamic range. Even at lower levels you still have more range than 16bit. It should only be an issue if you do something really dumb.
TheArchitect is offline   Reply With Quote
Old 30th July 2008   #98
Gear maniac
 
audibell's Avatar
 
Joined: Aug 2006
Location: New Orleans
Posts: 293

Quote:
Originally Posted by allencollins View Post
huh?

Anyone who worries about tracking levels in digital is clueless. As long as it's not clipping I DOESN'T MATTER.
Clueless!! preposterous!!! I, sir, fret and agonise over every last blip of my k-20 meters. And I NEVER clip, so how can I be clueless???
stuff and nonsense, mutter, mutter...
audibell is offline   Reply With Quote
New Reply New Reply Submit Thread to Facebook Facebook  Submit Thread to Twitter Twitter  Submit Thread to LinkedIn LinkedIn 



Thread Tools Search this Thread
Search this Thread:

Advanced Search

Similar Threads
Thread Thread starter Forum Replies Last Post
tracking levels in Control Room... Nik The Moan Zone 19 25th January 2006 08:08 PM
Digital tracking levels question copperx Low End Theory 10 10th November 2005 08:06 PM
Tracking levels? frist44 So much gear, so little time! 18 26th October 2005 05:49 PM
Console Channel levels vs stereo bus levels? Redsandblu So much gear, so little time! 6 4th May 2005 08:12 PM
TRACKING LEVELS doug_hti Expert Question & Answer Archives (read only archive, not open for new posts) 36 14th November 2002 12:14 AM


All times are GMT +1. The time now is 01:48 PM.

Home - Search Forum - Contact Us - Terms Of Use - Advertise on Gearslutz - All Advertisers - Archive - Top
 
 
Powered by vBulletin®
Gearslutz.com LTD - UK Company Number 7597610.
Registered Office - 35 Ballards Lane, London, N3 1XW.
Hosted by Nimbus Hosting.

SEO by vBSEO ©2010, Crawlability, Inc.