Your best home recording tips/advice/wisdom...

[Woodstock School Of Music]

Originally Posted by Joseph Hanna

Naw that's just not true Bob. Virtually every software (and 99.9%) of hardware units, that's of course including eq's, compressors, limiters, reverbs, delay's and well.... you name it, have input and output structures designed specifically to address that very situation.. If you're boosting frequencies in an eq section and didn't intend to increase the gain structure it's incumbent on the engineer to compensate at that stage. In this case the eq. Using the Pro Tools eq (as a recognizable example) there is a comparative input and output LED ladder stacked upon each other. At the end of the LED ladder there's both an input and output gain adjustment stage. The ladders are there for a quick glance at what the eq is doing to the gain stage comparative to the original. If a device introduces gain and you didn't intend for it to introduce gain...adjust it at the plug-in level. In fact in any DAW the single most easily overloaded devices are plug-ins. Digi has always preached the "control overloads most judiciously at the plug-in level. The same thing of course applies for subtractive eq. If you've altered the gain stage (in this case lowered) by inserting a subtractive eq setting, the correct move (if you didn't intend to lower the gain) is adjust it at the output stage of the plug-in. I'd never shy away from any technique (like additive eq) because it eats up head room. If in fact it does eat up head room the engineer ain't doin his/her job. In all of this mixing levels are best served at the "fader" stage of the gain journey and a savvy engineer will be fairly diligent to make sure any bus routes maintains that level of unity at grand efforts.


This goes hand in hand with the continuing mis-use of compressors and limiters. By and large (and particularly if there's a make-up gain circuitry built in) simply inserting a compressor can make the gain structure bump. The knee jerk reaction is always; things that are louder initially sound better. The correct use (at least most of the time) is to re-adjust the gain stage so the compressor does it's intended chore without adding the misleading component of "more gain".

In the end I'm adamantly opposed to the web forum wives tale mind-set that subtractive eq is the correct way for folks to use eq. That is patently false. The truth is a practical, sound, fundamental understanding of both gain staging and eq makes subtractive and additive eq an equally vital tool for anyone mixing audio.

Lots of good points here

Concerning "The knee jerk reaction is always; things that are louder initially sound better" this is 100% true and is very important for the engineer to be aware of that. Related to that would be when putting a compressor on a track match the level to the raw track so you can hear the two at equal levels and then A/B the bypassed vs non bypassed. It's real easy to get tricked if you call up a preset and it's instantly louder and bolder but if you pull it down to the same level as the raw track then you'll really hear what the compressor is doing. Just one example

Have more than one set of speakers to mix on if you can and when you mix listen to it at different levels including very quiet. Chris Lord Alge who mixes a majority of the Alternative and heavier bands mixes at a level so quiet that you can carry on a normal conversation while the music's playing. If you can get the music to pop or shine at a quiet volume it should be even better louder.

Another trick of mine is to let the mix play and walk into another room and do something like wrap cables, clean up or anything to divert your complete attention. Don't laugh but it's amazing what you can spot when you cut off the analytical side of your brain and listen from a casual perspective and distance.

When you mix something let it sit over night and listen in the morning. The first play with fresh ears should reveal if anything is not right

If your recording digital keep your levels low

Instead of dousing a track with compression to control the volume try using automation instead for general volume control and use the compressor for it's sound and back off the threshold/ratio

Don't automatically assume you have to slap compressors across everything

Don't get too caught up in the "If I only had this" syndrome, yes it's nice to have better gear but there is no magic bullet

None of this is the only way just some suggestions

[Joseph Hanna]

Originally Posted by The dman

Another trick of mine is to let the mix play and walk into another room and do something like wrap cables, clean up or anything to divert your complete attention. Don't laugh but it's amazing what you can spot when you cut off the analytical side of your brain and listen from a casual perspective and distance.

This! Extremely important especially in high stress/high concentration scenarios. I've long ago learned to program a "hot-spot" screen saver that gets triggered by dragging my mouse cursor to the lower left corner of my screen. It immediately triggers a placid screen saver that takes my "eyes" away from looking at wave forms. I use this all day, every day.

Originally Posted by The dman

When you mix something let it sit over night and listen in the morning. The first play with fresh ears should reveal if anything is not right

Yep. This is another one. In my world I don't EVER get a "tomorrow morning". The projects are mixed and shipped immediately. It absolutely causes a whole new mind-set to mixing and is a far more difficult task. That said and to prove that point, I've never worked on a television show that doesn't have some sort of postmortem meeting. It's usually in a room full of fellow editors and spear headed by some member of the executive producers staff with the intent to review the quality of the production. The show is played back as if we were home watchers/listeners. There have been times when I'm absolutely stunned beyond belief at how bad my mix was. Nothing ruins a day like a big-wig looking at you square in the eyes and asking "how can things have gotten this bad?" In the end when mixing audio (particularly under stress) we tend to latch on to certain things and completely ignore others. The more time between listening to mixes the undeniably better they become.

[dberch]

The only thing I'd like to add is this:

If you normally play and sing at the same time, that's the way you should record. Don't try to lay down the guitar track first then overdub the vocals. At least in my case, I had no idea of how much I subtly alter the guitar part as I'm singing to go long with the vocals. I just could not get vocals to match the guitar part when trying to do vocal track. Had to do it over and record both simultaneously.

My apologies if this has already been stated above. I didn't take time to read everything...

David

[Woodstock School Of Music]

Originally Posted by dberch

The only thing I'd like to add is this:

If you normally play and sing at the same time, that's the way you should record. Don't try to lay down the guitar track first then overdub the vocals. At least in my case, I had no idea of how much I subtly alter the guitar part as I'm singing to go long with the vocals. I just could not get vocals to match the guitar part when trying to do vocal track. Had to do it over and record both simultaneously.

My apologies if this has already been stated above. I didn't take time to read everything...

David

Some people are bothered by that more than others but I agree if you can't cop the feel you like doing the guitar and vocal individually then record them together.

I have one client that's done a few cd's with me and he has always recorded his guitar and vocals together, he's a fish out of water doing it the other way. It would make it easier on both of us to record separate tracks but if the music suffers it's not worth it.

[Bob Womack]

Well, Joseph, I was trying to help new guys. I didn't say subtractive EQ is the only way to go. What I did said was it is the natural tendency of someone new to the field to use additive EQ when subtractive may work better. I also said that additive EQ uses up headroom. This is true. If you do additive EQ and the level for your sound or track goes up, you have to pull the overall level back in order to bring it back in. As a result, the overall level of any broadband sound has just been dropped. The opposite is true if you have a bulge in the frequency response of a broadband sound at some point and you tame it with subtractive EQ - you are often able to bring up the sound in the mix as a result. Also, with additive EQ, you can add a frequency spike to a sound that adds loads of local gain but not much of loudness within a mix. Those frequency-dependent spikes can also not be detected well when you are using simple metering (rather than some form of full spectrum metering, allowing you to end up with a level that looks like it is correct while your EQ spike is saturating the next signal stage at its frequencies only.

Let me give you an example: In the 1980s my TV network suddenly began getting reports that our VOs were distorted and slurpy sounding. We began blueprinting our signal chain to find the distortion. The mic was a Neuman U-47 that had just come form a rebuild at Gotham Audio. Its signal went to a Neve 8024 console (with 1073 preamps, bass rolled off), with a Neve compressor inserted (I think a 2264) and 3-6 db of peak limiting used. Throughout the console signal chain the available 20db Neve headroom was maintained. The signal was recorded to an Ampex ATR-102 adjusted for flat (not hyped) response at +6 (320 nanowebers per meter) on Ampex 456 tape. From there the VOs were copied real-time to a calibrated broadcast cart machines, sneaker-netted to master control, and played off an identical calibrated broadcast cart machine. We found that the signal was clean all the way out our satellite transmitters and metered out at the same conservative 0VU.

However, our satellite return was indeed distorted and slurpy. We eventually discovered that some of the existing analog satellite technology had a real problem: While the gain structures of some of those earlier satellite transponders were designed for the signal to run at 0VU, they only offered four db of headroom above 0 before distortion. Right before we started to get the reports, my network had switched satellites and ended up on one of those low-headroom transponders. For several years we had been doing our male voice-overs for TV with a Neumann U-47 microphone. Those who know the mic know that it has an intentional bubble in the upper-midrange designed to make a voice more intelligible when you use the single mic to pickup a solo voice and orchestra. It is part of what defines the rich sound of the mic. While the overall signal metered out just fine at a conservative 0VU, that bulge in upper mids pushed the satellite transponder into distortion right at those frequencies only. When we put a real time analyzer across the master control output and looked at the VOs, that bulge above OUV was exactly what we saw. We contacted the satellite provider and THAT's when we had the headroom issue revealed to us. We solved the problem temporarily by feeding the mic white noise, finding the response bulge with a real-time analyzer, and dialing it out with an EQ. We found a more permanent solution in switching to a mic that didn't have the bulge. And eventually we moved off that satellite for good and could use whatever we wanted.

It is normal for a new guy to hear bulges and try to correct by boosting the adjacent frequencies with EQ. It is possible to end up with spikes that don't show unless you are looking at them with spectrum metering. However, most home recordists don't start off with spectrum metering. While teaching audio classes I've found it to be easier to start them out watching for frequency build ups and learning the artof subtractive EQ. In the west there seems to be a fear of backing off anything. Have you ever noticed that people prefer to change lanes to get off the Interstate by zipping up in front of you and then pulling across your bow rather than dropping back to that huge space behind you? It's the "more is better than less" syndrome. You see it in mastering too... Hehehehe...

Oh, and by the way, I use additive EQ all the time, but I try to avoid spikes and frequency built-ups.

Cheers,

Bob

Originally Posted by Joseph Hanna

Naw that's just not true Bob. Virtually every software (and 99.9%) of hardware units, that's of course including eq's, compressors, limiters, reverbs, delay's and well.... you name it, have input and output structures designed specifically to address that very situation.. If you're boosting frequencies in an eq section and didn't intend to increase the gain structure it's incumbent on the engineer to compensate at that stage. In this case the eq. Using the Pro Tools eq (as a recognizable example) there is a comparative input and output LED ladder stacked upon each other. At the end of the LED ladder there's both an input and output gain adjustment stage. The ladders are there for a quick glance at what the eq is doing to the gain stage comparative to the original. If a device introduces gain and you didn't intend for it to introduce gain...adjust it at the plug-in level. In fact in any DAW the single most easily overloaded devices are plug-ins. Digi has always preached the "control overloads most judiciously at the plug-in level. The same thing of course applies for subtractive eq. If you've altered the gain stage (in this case lowered) by inserting a subtractive eq setting, the correct move (if you didn't intend to lower the gain) is adjust it at the output stage of the plug-in. I'd never shy away from any technique (like additive eq) because it eats up head room. If in fact it does eat up head room the engineer ain't doin his/her job. In all of this mixing levels are best served at the "fader" stage of the gain journey and a savvy engineer will be fairly diligent to make sure any bus routes maintains that level of unity at grand efforts.


This goes hand in hand with the continuing mis-use of compressors and limiters. By and large (and particularly if there's a make-up gain circuitry built in) simply inserting a compressor can make the gain structure bump. The knee jerk reaction is always; things that are louder initially sound better. The correct use (at least most of the time) is to re-adjust the gain stage so the compressor does it's intended chore without adding the misleading component of "more gain".

In the end I'm adamantly opposed to the web forum wives tale mind-set that subtractive eq is the correct way for folks to use eq. That is patently false. The truth is a practical, sound, fundamental understanding of both gain staging and eq makes subtractive and additive eq an equally vital tool for anyone mixing audio.

[Ivan Lee]

- Get a good set of closed back headphones for tracking
- With just mic choice and position (maybe some verb where latency doesn't matter) get a sound in your headphones that sounds great and inspires you
- Look up "gain staging" - see how it applies for your situation
- Record into a dead corner (as simple as two moving blankets) over into an untreated room
- Learn Cardioid - figure8 - and Omni patterns. Always bear in mind to use the most appropriate.
- Learn proximity effect - close is good - but too close may not be
- Done always better than perfect
- Keep AC lines and Signal lines (especially Mic Lines) separate. A cheap solution is to use the foam pipe liners from Home Depot to isolate the AC lines in particular. Where you must cross power and signal lines use 90 degree intersections - don't run parallel.
- Look up star ground - use it.
- If this "techie" stuff doesn't appeal to you - $20 an hour will easily get you someone to record you instead (with most likely $10K worth of gear!). In my neighborhood $50/hr will get you near world class !!

WOW - my 500th post :-)

[Joseph Hanna]

Originally Posted by Bob Womack

If you do additive EQ and the level for your sound or track goes up, you have to pull the overall level back in order to bring it back in. As a result, the overall level of any broadband sound has just been dropped.

Yes..that's my point. I agree. If the creative choice dictates the use of eq and that choice raises the gain structure ostensibly by additive eq (and the gain structure wasn't intended to be raised) then the only logical choice is to lower the output of the eq in order to a) affect the changes intended by using the eq and b) maintain and control a manageable gain structure.

Originally Posted by Bob Womack

The opposite is true if you have a bulge in the frequency response of a broadband sound at some point and you tame it with subtractive EQ - you are often able to bring up the sound in the mix as a result.

I mean sure??!? Logically if you reduce the gain via an eq you've effectively increased the headroom of the track and on some small level the session. That's a tough one to contest and kinda obvious. But (particularly with plug-ins) the most effective place to minimize gain changes, that have arisen out of instantiating an eq, is at the source, which is of course, the eq itself.

Originally Posted by Bob Womack

Also, with additive EQ, you can add a frequency spike to a sound that adds loads of local gain but not much of loudness within a mix.

Yea I mean sure. If you have a source sound that's peaky and 1k heavy and you add an eq that boosts 1k even more, somethings gonna happen and it obviously could lead to peaks. I guess the assumption is even the most rudimentary audio guy would realize that sooner than later. Still if the creative choice was to actually bump 1K in a heavy 1k track (and that's of course realistic for something) the first place to start to get the peaks under control and regain headroom, would be the eq.

Originally Posted by Bob Womack

Those frequency-dependent spikes can also not be detected well when you are using simple metering (rather than some form of full spectrum metering, allowing you to end up with a level that looks like it is correct while your EQ spike is saturating the next signal stage at its frequencies only.

Yea i don't really get this. You can obviously have a build up of frequencies in any mix. That of course being multiple tracks that are providing fundamental as well as overtones buildups around like and similar frequencies. It's also true that a typical peak/rms meter won't shed much accurate information as to the source of those buildups particularly while monitoring the mix as a whole. I'm assuming you're referring to this as peaks. I think of peaks as something more along the line of a spike in dB level than I do a buildup causing a disproportionate output at a given frequencies. But again it's kinda fundamental as to the issue of additive eq that if you add gain in a particular frequency range (particularly if that frequency range is already hot) it's gonna cause problems. That's something I'd guess most (even beginners) would recognize. That said even if that mistake was made if the proper steps were taken to reduce the gain stage AT the eq there'd be no (or at least insignificant) "peak" changes, at least on a track by track basis.

Originally Posted by Bob Womack

Let me give you an example: In the 1980s my TV network suddenly began getting reports that our VOs were distorted and slurpy sounding. We began blueprinting our signal chain to find the distortion. The mic was a Neuman U-47 that had just come form a rebuild at Gotham Audio. Its signal went to a Neve 8024 console (with 1073 preamps, bass rolled off), with a Neve compressor inserted (I think a 2264) and 3-6 db of peak limiting used. Throughout the console signal chain the available 20db Neve headroom was maintained. The signal was recorded to an Ampex ATR-102 adjusted for flat (not hyped) response at +6 (320 nanowebers per meter) on Ampex 456 tape. From there the VOs were copied real-time to a calibrated broadcast cart machines, sneaker-netted to master control, and played off an identical calibrated broadcast cart machine. We found that the signal was clean all the way out our satellite transmitters and metered out at the same conservative 0VU.

However, our satellite return was indeed distorted and slurpy. We eventually discovered that some of the existing analog satellite technology had a real problem: While the gain structures of some of those earlier satellite transponders were designed for the signal to run at 0VU, they only offered four db of headroom above 0 before distortion. Right before we started to get the reports, my network had switched satellites and ended up on one of those low-headroom transponders. For several years we had been doing our male voice-overs for TV with a Neumann U-47 microphone. Those who know the mic know that it has an intentional bubble in the upper-midrange designed to make a voice more intelligible when you use the single mic to pickup a solo voice and orchestra. It is part of what defines the rich sound of the mic. While the overall signal metered out just fine at a conservative 0VU, that bulge in upper mids pushed the satellite transponder into distortion right at those frequencies only. When we put a real time analyzer across the master control output and looked at the VOs, that bulge above OUV was exactly what we saw. We contacted the satellite provider and THAT's when we had the headroom issue revealed to us. We solved the problem temporarily by feeding the mic white noise, finding the response bulge with a real-time analyzer, and dialing it out with an EQ. We found a more permanent solution in switching to a mic that didn't have the bulge. And eventually we moved off that satellite for good and could use whatever we wanted.

It is normal for a new guy to hear bulges and try to correct by boosting the adjacent frequencies with EQ. It is possible to end up with spikes that don't show unless you are looking at them with spectrum metering. However, most home recordists don't start off with spectrum metering. While teaching audio classes I've found it to be easier to start them out watching for frequency build ups and learning the artof subtractive EQ. In the west there seems to be a fear of backing off anything. Have you ever noticed that people prefer to change lanes to get off the Interstate by zipping up in front of you and then pulling across your bow rather than dropping back to that huge space behind you? It's the "more is better than less" syndrome. You see it in mastering too... Hehehehe...

Holy cow...ya kinda lost me here Bob Lot's of fancy words that are kinda over my head. "Sneaker-netted" and "nanowebers" sound cool though It appears that you're saying the satellite feed process was not encoding certain frequencies properly. That there was a frequency eq build up at the peak point of the Neumann, yet no overs 0dB, and that for some reason and at some point in that process the VO became sloppy, or slurry?? That's a frequency problem not a gain structure problem yes??

Anyway I deal with satellite feeds every day and to every primary affiliate on the planet. All sorts of things happy in that process. What the feed company does with it and worse what local affiliates do with is anybody's guess and I've heard some genuinely crazy results. It also appears that you solved that problem, at least initially, by finding the offending frequency and eq'ing it out. If that's the case...I get it

Originally Posted by Bob Womack

I try to avoid spikes and frequency built-ups.

Me too Bob, me too.

[ombudsman]

Originally Posted by Joseph Hanna

Yea I mean sure. If you have a source sound that's peaky and 1k heavy and you add an eq that boosts 1k even more, somethings gonna happen and it obviously could lead to peaks. I guess the assumption is even the most rudimentary audio guy would realize that sooner than later.

I think it's fair to say that in the era of the cheap digital home studio, audio guys have gotten a lot more rudimentary than they used to be.

A lot of people doing recording don't have a clear understanding of gain structure, and to be fair sometimes these things can be hard to track down. If you tracked too hot (as a lot of beginners will do) you can run out of headroom in your plugins at some frequencies without seeing any clipping at the channel or buss, and yet you're hearing it.

I really love the analyzer in the same window feature of the EQ plugins in Logic. If all EQs had that we would be a lot better off.

Originally Posted by Joseph Hanna

It appears that you're saying the satellite feed process was not encoding certain frequencies properly. That there was a frequency eq build up at the peak point of the Neumann, yet no overs 0dB, and that for some reason and at some point in that process the VO became sloppy, or slurry?? That's a frequency problem not a gain structure problem yes??

He's talking about running out of headroom in an analog signal, clipping at +4dbvu.

[Joseph Hanna]

Originally Posted by ombudsman

I think it's fair to say that in the era of the cheap digital home studio, audio guys have gotten a lot more rudimentary than they used to be.

A lot of people doing recording don't have a clear understanding of gain structure, and to be fair sometimes these things can be hard to track down. If you tracked too hot (as a lot of beginners will do) you can run out of headroom in your plugins at some frequencies without seeing any clipping at the channel or buss, and yet you're hearing it.

I absolutely agree. Where I might have misunderstood Bob's original post it is for the very reasons you mentioned (that is a lot of folks don't have a clear understanding of gain structure) that I get sideways a bit when forum wives tales begin to pop up. The topic of subtractive eq is mentioned here and mentioned often. My point (and I think Bob will agree) is, rather than promote the use of subtractive eq a better suggestion would be to promote the idea of sitting down and taking a stab at grasping the gain-stage fundamentals throughout the entirety of any and all buss structures within the context of a mix.

Originally Posted by ombudsman

I really love the analyzer in the same window feature of the EQ plugins in Logic. If all EQs had that we would be a lot better off.

Blue Cat Audio has an intensive analyzer that's freeware. I think it's available in all primary protocols. I do monitor an analyzer during my television mixes but I've learned to only peripherally follow it. As with anything in audio, sometimes the meters play cruel hoaxes. With "long form" mixing (which is the the show itself) analyzing meters are mostly ineffective as the audio (particularly the cues) are already mixed by whoever is composing for the show. The SOT's (generally dialog) changes quality-wise depending on the show but usually these days really well recorded. In the end it's mostly a matter of levels and a general feel. Analyzers just aren't critical here. The promo's however are a smash-fest and tryin to squeeze in all the elements the producers demand is a task. An analyzer becomes crucial.

Originally Posted by ombudsman

He's talking about running out of headroom in an analog signal, clipping at +4dbvu.

Well yep I knew he was speaking of analog but he indicated he had not exceeded 0dBvu. That of course would indicate (all things being calibrated, and by the terms he was tossin around I gotta assume the room was calibrated I'm still tryin to find a way to work in "sneaker-nettted" into a conversation with the guys in the other audio bays here) if there was distortion on the VO's it was either imbedded on the track or it was a function of the encoding. That is to say if the gain structure was below the point of clipping 0dBvu or 0dBfs it's not likely to be the culprit at hand

[ombudsman]

Originally Posted by Joseph Hanna

Well yep I knew he was speaking of analog but he indicated he had not exceeded 0dBvu. That of course would indicate (all things being calibrated, and by the terms he was tossin around I gotta assume the room was calibrated I'm still tryin to find a way to work in "sneaker-nettted" into a conversation with the guys in the other audio bays here)

not exactly, he said it showed as 0dbvu on the meter but they were used to a clipping point above +4.

Originally Posted by Joseph Hanna

if there was distortion on the VO's it was either imbedded on the track or it was a function of the encoding. That is to say if the gain structure was below the point of clipping 0dBvu or 0dBfs it's not likely to be the culprit at hand

My point in that comment is that there was no encoding since it was analog all the way through in the 80s. The gain was nominal 0 dbvu on the meters but it exceeded +4 in the mids and clipped. It's a gain structure problem by virtue of the same type of signal working OK before and after they were on that particular satellite which was uncharacteristically low headroom.

[louparte]

Originally Posted by bobby b

Advice......hmmmm

Dont mix with headphones.

Disagree with that tip. Best mix I ever made was on K240's in an airport using Auria on my iPad.
I've made good mixes on HD280's too. But I always A/B'd them w/my monitors. The airport mix I had nothing but the K-240's.

[Bob Womack]

Joseph, my point in the first post was aimed at new operators who might not be able to recognize that their problem is a boosted range of frequencies that are masking an otherwise fairly flat response. I find that, coupled with the accelerator pedal phenomenon mentioned above (always more instead of less), many new operators tend to rush to boost rather than cut. I was merely suggesting that they explore the negative side of the EQ in an attempt to stem the natural tendency I've observed in new guys to cure all ills with additive EQ.

The point of the whole signal chain thing in the second post was to show how we proofed our plant from one end to the other and were able to show that a clean signal that metered at a conservative 0VU was still able to cause distortion in a low-headroom device down the line in one frequency band only. In other words, 0VU indicated was averaged over all frequencies and masked a large spike at just a few frequencies generated by the microphone. That applies here because these days we meter in a zero-headroom manner using peak meters that terminate at the end of headroom. In way that is a parallel to the example, someone not accustomed to thinking about spectrum of a signal can EQ it so that it looks fine on a peak meter but is still capable of distorting the chain ahead of him.

And since you asked, "Nanowebers/meter" is the measurement of fluxivity in analog magnetic recording. Its the measure of how hard you are pushing the tape with magnetic force while recording. Ampex 456 tape was spec'd from the factory at +6 (320 nanowebers/meter) above the 1960s standard of 185 as 0 tape level. My attempt there was to point out that the machine was set-up conservatively for itself and the tape used.

"Sneaker-netting" is the process of physically picking up a piece of media and walking or running it to the next place in the signal chain. The term is a humorous and ironic way of describing the old-time business of picking up the reel, cassette, cart, hard drive, etc., containing the program and carrying it to the next place it is going to be used. It was standard practice before the industry went to LAN-based delivery and is still practiced in some plants that deal in large-scale production, such as film post houses. Example for you: "Sneaker-net this work order over to Sam, please." Thus, you can communicate you wishes and sneak in a little good-natured humor as well. That is, of course, if you use physical work orders. We've gone non-linear there, too.

All things in balance. I think we agree here. And I respect your experience.

All the best,

Bob

[Soaranator]

Get the best stuff you can afford. Use a tuner. Use a click track.

[Joseph Hanna]

Originally Posted by Bob Womack

Joseph, my point in the first post was aimed at new operators who might not be able to recognize that their problem is a boosted range of frequencies that are masking an otherwise fairly flat response. I find that, coupled with the accelerator pedal phenomenon mentioned above (always more instead of less), many new operators tend to rush to boost rather than cut. I was merely suggesting that they explore the negative side of the EQ in an attempt to stem the natural tendency I've observed in new guys to cure all ills with additive EQ.

That is great advice Bob! When you put it in this context it's instructive and informative to the novice-ish folks I think we both wish to reach. I do understand the engine behind your suggesting subtractive eq. Unfortunately it's a subject that's already an internet wives tale heavy weight meme and without any explanation attached (that's is to say someone just advocating the phrase "subtractive eq") often fuels the fire!!

Originally Posted by Bob Womack

The point of the whole signal chain thing in the second post was to show how we proofed our plant from one end to the other and were able to show that a clean signal that metered at a conservative 0VU was still able to cause distortion in a low-headroom device down the line in one frequency band only. In other words, 0VU indicated was averaged over all frequencies and masked a large spike at just a few frequencies generated by the microphone. That applies here because these days we meter in a zero-headroom manner using peak meters that terminate at the end of headroom. In way that is a parallel to the example, someone not accustomed to thinking about spectrum of a signal can EQ it so that it looks fine on a peak meter but is still capable of distorting the chain ahead of him.

Yea I think I get where you're headed here. As you know the post industry radically change delivery specs. It makes for a much easier day as by and large the spec is relaxed comparatively and in so I don't have to hug the meters 12 hours a day. BUT...in years past we micro-managed mixes, particularly large mixes. We did so by painstakingly conforming omfi files into stems. Those stems would usually include SOT's, SOT's audience, SOT natural sounds, any FX's, narrative and or VO, additive laugh and finally music. We micromanaged each stem by first going in and setting the dB levels of each stem by a predetermined level. EVERY syllable was set to within a dB or two of one another. From there we mixed. The reason however we did this was that rouge "over". As I mentioned earlier we mix to feed. That is to say most days I mix until a PC tells me it's 10 minutes to feed. If I realize in that 10 minute countdown i've got rouge "overs" I'm screwed. I don't know the spec in your area of the woods but here in Los Angeles if your feeding the network affiliates the companies responsible for the satellite feed "reviews" the mix. If there are "overs" they won't take the responsibilities of feeding the show. A network show that doesn't feed on pre purchased satellite time means....I lose my job. So in short I get what undetected overs mean in the general scheme of things.....intimately

Originally Posted by Bob Womack

And since you asked, "Nanowebers/meter" is the measurement of fluxivity in analog magnetic recording. Its the measure of how hard you are pushing the tape with magnetic force while recording. Ampex 456 tape was spec'd from the factory at +6 (320 nanowebers/meter) above the 1960s standard of 185 as 0 tape level. My attempt there was to point out that the machine was set-up conservatively for itself and the tape used.

Good to know...never been involved in the analog calibration processes.

Originally Posted by Bob Womack

"Sneaker-netting" is the process of physically picking up a piece of media and walking or running it to the next place in the signal chain. It was standard practice before the industry went to LAN-based delivery and is still practiced in some plants that deal in large-scale production, such as film post houses.

Ah-ha! One of my very best friends and former co-worker moved into mixing trailers at Sony Post down in Culver City. You can bet the next time I talk to him I'm gonna work sneaker-netting into the conversation!!

Originally Posted by Bob Womack

All things in balance. I think we agree here. And I respect your experience.

Like-wise

[Bob1131]

There's been some very good and interesting tips posted so far. As a complete novice home recording hobbyist, the two most important tips I have gotten are:

1. Never mix with tired ears, and

2. Use a "reference" recording to compare to when recording and mixing your tracks. Select a recording that you like the sound quality of and try to get as close to that sound as possible both when recording and mixing. I multi-track with guitar, vocals, bass, keys and drums but, sometimes it is only guitar and vocal. If my recording has the full band sound I'll use a full band reference recording like the Eagles, America, or a newer group. If it is just a solo guitar and vocal I'll use a Jack Johnson or similar type of track. When setting up, I a/B my set up to the reference, and when I mix the final track I adjust EQ (add or subtract as needed) to get as close to the reference track sound as I can, again A/B'ing after each adjustment.