This question came in from Joseph K, who asks:

”Hey Russ & Keith, I am a newby Pro Tools 9 user and I find your site very helpful. I work with Virtual instruments mainly. I am concerned as to when I should convert instrument tracks to audio tracks. Can Plug-in Instrument tracks be mastered just as well as audio tracks or do they need to be converted to audio tracks within ProTools before Final mixing and mastering? Not sure if I should be wasting too much time converting all instrument tracks to Audio if it’s not really neccessary..

Thanks for all of your help, I am very appreciative for all you do to help us artists.”

Good question, Joseph! I get asked this relatively often and there’s one definitive answer, in my opinion and that is to always bounce your VI tracks.

The reason I say that is this - should you not open that session for a long time after finishing it, but then need to open it to do a remix or instrumental version or any number of other reasons, your plug-ins and VI’s may have changed since you last opened it, causing delays in being able to complete the new version. Think ahead - you could find yourself needing another mix of that song a few years down the line. If your OS has been upgraded, or support for various plug-ins ended since you opened it you may find yourself in a difficult position of the song not sounding the same, or lacking instruments that no longer work.

While support for plug-ins and VI’s is usually not a problem, over the years I’ve seen them dropped from developers rosters. If you created a song with those instruments and didn’t bounce them, but you’ve gone through an OS or DAW upgrade or two, you may find they don’t work on your current system. In that case you really only have a few choices - try to recreate the sound or revert to an older OS and/or DAW version that still supports those plug-ins.

If, however, you had the foresight to bounce your VI as audio - you’re good to go no matter what.

With Pro Tools you may find you need to bounce your VI’s while working anyways, to free up processor load and avoid errors. Avid has reassured us that VI performance enhancements will be coming to Pro Tools at some point but, until that happens, the current performance level may not meet your needs for larger compositions and bouncing may be necessary.

As part of my routine I will always bounce all VI’s to audio tracks after a song is finished. There are relatively easy ways to bounce multiple tracks simultaneously, reducing the time needed to do so over bouncing them one by one. The most efficient method is to use a bus based bounce on all VI tracks to a new audio track. Set up as many new audio tracks as you have VI’s to bounce, then route the output of each VI to their respective new audio track. Record enable the new tracks and away you go - bouncing multiple VI’s will only take as long as the length of the song.

After bouncing you can disable the VI’s to free up processor load and hide them to keep them from cluttering up your Edit or Mix windows.

For the record - there is NO difference between a VI track and one that’s been recorded as audio. Either would work fine to create your final mix and prepare it for mastering. So, while it’s not *necessary* to bounce them before your final mix, doing so for the reasons above is a good idea at least for archival purposes.

I hope this has been helpful!

- KC

Hello and welcome to ‘Ask Keith’ on the Air Users Blog. I’ll be busting my AUB cherry today with a little tech perspective on analog vs. digital or vinyl vs. CD. This question came from Anthony Mena (aka ScrewMaker) - if you don’t recognize the name, he’s a monster audio engineer and composer, having worked with Nine Inch Nails and Filter, amongst others, and contributed to some of the most popular sample and sound effects libraries from Big Fish, MVP, Producer Loops/Boss Loops and more. To top it off he’s from the same city I was born in - Janesville, Wisconsin. Coincidence? I think not.

Today we’ll be talking not about which is better or worse per se, but about the technical aspects of each and how they relate to our perception of ‘better’ and how understanding the differences can improve your production. Let’s dive into the technical aspects where vinyl fails horribly.

You’ve probably already read endless debates about why vinyl or analog sounds better than CD - it rages both ways, with some convincing and not so convincing arguments on either side. I don’t plan on putting this debate to rest because we all know how much people like to ignore facts and state their opinions on the internet. What I’ll try to do here is separate the opinions from the facts.

Fact: Vinyl has less dynamic range than CD. Vinyl clocks in at about 75db dynamic range, while CD’s have about 96db. Analog tape, with appropriate noise reduction units like Dolby, get to about 80db. So - more dynamic range = better, oui?

Not so fast - remember how, in audio production, even a 1db difference in listening volume can cause someone, when listening to the *exact* same song, to perceive the louder one as being ‘better? With a more limited dynamic range vinyl will often be interpreted as ‘better’ sounding, as the quieter parts won’t be as quiet as the same song on a CD.

But, wait a minute you say, couldn’t I just compress my mixes to achieve an overall lower dynamic range on the CD, thus increasing it’s perceived volume? Indeed, you can - and this is where the loudness wars really began.

For a little historical perspective, let’s look at what went into creating a vinyl record. Vinyl has inherent limitations based on the medium. I’m sure you understand how vinyl records are made already (if not, there are plenty of articles already covering this topic - GIYF!). When we look at things like the RIAA curve (standardized in 1954) we begin to understand the limitations of vinyl - the RIAA curve was created to overcome some of those limitations. The RIAA recording curve has a 20db cut in the bass with an almost straight line through the EQ spectrum from 20hz to 20khz up to a 20db boost in the highs, with a slight ‘bump’ in the 500hz range. This was developed to address issues on both ends of the frequency spectrum - in the bass range it allowed for grooves to be cut closer together, resulting in longer playing times, and, most importantly, to reduce the physical deviation that a stylus had to navigate to reproduce bass. Because vinyl is analog, cutting a vinyl record without any EQ cut in the bass would result in excessive distortion as the stylus tried to follow a much wider groove. On the high end the boost in higher frequencies would, upon the opposite EQ curve applied during playback, result in lower noise on the high end. If we boost the highs during cutting, we increase their level in relation to the noise floor - for instance, if we have a high end that’s 40db above the noise floor and cut it without any EQ adjustments we get just that - 40db of dynamic range above the noise floor. If, however, we boost it 20db during cutting, increasing it’s level another 20db above the noise floor, upon playback we now have 60db of dynamic range above the noise floor (remember the RIAA EQ curve cuts the high end 20db during reproduction - both the signal and the noise floor are reduced in the high end at that stage, but the ratio remains the same.).

So, how does this apply to the loudness wars? During the development of vinyl reproduction engineers had to find ways to get the best sound quality and still fit it within the dynamic range of the medium itself - thus advancements in compression and EQ were invented to make that possible. Later down the line we start to see how engineers took the information they understood about ‘perceived loudness’ and applied it during the mastering process. Remember what I said about even a 1db difference in volume being perceived as ‘better’? The loudness wars took that to the extreme - if 1db difference is better, 10db difference is 10x better, right? So many have come to believe, to their detriment. We’ll revisit why in a bit.

Fact : Vinyl has a lot of crosstalk, to the tune of about 30db of total channel separation. CD/Digital, on the other hand, has no crosstalk to speak of. CD’s should thus sound better, yes?

Again, we have specs that say one thing, but ‘real world’ examples show otherwise. Let’s take a look at that ‘low’ channel separation number for vinyl. 30db doesn’t seem like much, but let’s look at what that really means. If you’ve done your homework you’ll know that a 10db difference in volume equals ‘twice’ as loud (or soft) to the human ear. We’re going to lithely skip over any reference to the Fletcher Munson curve at this point for simplicities sake. So that 30db of channel separation, if we were to express it as a ratio, would mean the crosstalk is 1/8th as loud in the opposite channel, or the difference between sitting in your living room and running your lawn mower. So, while more channel separation, on paper, looks better - the way we perceive it doesn’t mean we’ll actually hear much of a difference.

Fact: Vinyl has more distortion than CD’s. CD’s have a THD+N (Total Harmonic Distortion + Noise) of well under .01% (using quality electronics). Vinyl, by comparison, treats us to THD+N specs upwards of 7%. Digital CLEARLY will sound better!

This is another area where specs can be stated and yet clearly not tell the whole tale. Often what’s talked about most is the type of distortion in each medium. Digital has almost zero in the way of distortion and even the claims of it’s ‘undesirable’ 3rd order harmonics are overblown as they typically fall well outside our perceptual range and are so low in comparison to the average signal level as to be non-existent (we won’t even touch the debate of whether 3rd order harmonics are really ‘bad’ or not…there’s enough there for a whole article in itself! Actually, here’s a link to a great article the blows the 2nd and 3rd order harmonics mumbo-jumbo to bits: http://www.stereophile.com/reference/406howard/index.html). Vinyl has a much higher distortion, period. However, much like the crosstalk, it’s perceptual volume is much lower than the signal - typically to the tune of 40db lower, which is a level 1/16th as loud as the signal. You’ll often see graphs of sine waves showing the 2nd or 3rd order harmonics with vinyl and digital, but a sine wave is not music - complex waveforms are very different from sine waves. Complex waveforms will invariably introduce BOTH 2nd and 3rd order harmonics in almost every medium and each component in the signal chain introduces it’s own harmonics, contributing to the overall distortion. On that note - while vinyl has more distortion, our perception of it, and it’s overall level vs. the source signal, is such that we don’t necessarily notice it or, if we do, it’s not a huge factor. This is one where personal preference plays a huge part and is more a factor of our unique listening apparatus (ears and brains) than anything. Just like some prefer metal and others easy listening, some with prefer the sound of vinyl, other CD’s.

The bottom line here - vinyl has ample dynamic range for the human ear and it’s distortion characteristics are such that they are perceived by many as being either unnoticeable or acceptable.

Now we’ll get to the meat of this by circling back to the loudness wars. We’ve all heard about it and have seen the horrible results. In my previous experiments with people, I’ve concluded it’s NOT a vinyl/analog vs. CD/digital debate, but rather a ‘how well was this recorded and mastered’ debate. In a previous life I had done a few listening tests with people - by recording a vinyl edition of a release to CD and having the subject listen to that vs. a recent reissue of the very same release directly from the remastered CD. I also took a few tracks from a CD and ‘pushed’ their overall loudness and compressed the dynamic range and threw those in the mix. The result - invariably the listener chose the version that had more dynamic range and less compression. The comment most heard was ‘I had to turn this one up a bit to get the same volume’. Indeed, they would have for the original, as it’s RMS volume level was notably lower than the remastered version. The other telling point was the vinyl version versus the CD version - the CD versions had been ‘remastered’ with the same reduction in dynamic range and increase in overall loudness. You would think that ‘louder = better’, but that’s been proven to not be the case.

The bigger picture shows that, whether it’s vinyl or CD has almost NO direct correlation to better or worse - it’s the dynamic range and overall loudness that is the biggest factor. So, our perception of vinyl being better (as stated by many) isn’t that vinyl is *actually* better, but that it offered more dynamic range because engineers weren’t yet trying to crush the living daylights out of it to ‘compete’. I bet if we threw some of the currently ‘remastered’ versions of records, or current releases that have been banging the top of the dynamic range spectrum, on vinyl the vinyl professors would say ‘vinyl sucks!’.

The biggest mistake being made in the loudness wars is the belief that a release mastered at lower levels and with more dynamic range isn’t going to ‘stand up’ to other releases. What they’re missing is that almost every listening medium has it’s own compression built in - radio, for instance, has compressors, leveler and limiters that push the sound into the FM radio dynamic range of under 50db and keep the overall dynamic range within a very tight spec so the listener experience is such that they don’t have to touch their volume control. iTunes also has built in dynamic range leveling and most consumers don’t even realize it’s there - they have it on by default. It won’t matter much how hard you push your masters - other than, if they have no dynamic range, everything is going to sound like it’s at the same volume, with no real ‘quiet’ parts or those lovely ‘big’ sounding choruses. If the chorus is at the same perceived volume as the verse, it’s going to have less of an impact.

What does this mean for your production? Simply put - don’t fall prey to the myths of the medium. Don’t fall prey to thinking you have to be ‘as loud’ as the other guy. Ease up on the compression and let your mixing and mastering reveal some true dynamic range. I, very recently, witnessed a musician slipping into this trap, so I know it’s still happening. I’m almost dreading hearing what the final master will sound like - I’m glad I got the one that won’t be squashed to death!

There’s a lot more history and technical talk that I could get into, especially regarding the developments in the early era of sound recording, but I think I’ve given enough meat for you to chew on for now. If you would like to talk more about this or have me cover this subject further, please leave a comment!

Cheers!

KC

Hello and welcome to my little corner of the AUB! First off - a HUGE thank you to everyone that's shown their support since my departure from Avid. I'm humbled beyond words. You guys are like my second family. Don't make that weird.

I'll try to keep this first post short and sweet, as I've got some serious relaxation to catch up on. Well, if you call working on a resume 'relaxation'.

Here's how this is going to roll - since this is a new endeavor for Russ and I, I'm sure it will need tweaking as we go along, so bear with us. Or else.

What this is: A place to pose a question that you've been unable to find a solution for on the DUC or after doing troubleshooting yourself. A place to ask technical question unrelated to troubleshooting that you would like to 'know more about'. I'll be straight up - I don't know the answer to everything and I will tell you when that's the case. What I *will* do is try to point you to resources that can fill in the blanks, when possible. I'm hoping to see good questions about workflow, too - the how and why to do a certain thing and how to make the process faster, easier and better.

What this is not: A place to get help that is freely and easily available elsewhere. A place where your question has been answered gobs of times on the DUC or by doing a Google search. I simply won't have the time to get looking at Sandra reports and log files to solve your problem - unless it's a *really* interesting one that  I'd consider a challenge.

What I'll be doing here is, rather than just telling you how to solve the problem, I'm going to give you a 'tech perspective' of the factors involved in the problem and how to think like a tech, so you can not only solve it yourself, but help Grandma out when she's having issues trying to record that overdub for her Christmas video.

Now that we all know the rules, let's go break some!

Lastly I want to thank Russ for this opportunity and his unflagging and unfettered support of the Pro Tools community. He's a rare breed indeed.

And a 1, and a 2…GO!