My - insert Sampler Based plug-in here (Superior Drummer, Groove Agent, etc) - is causing my DAW to

• crash
  
• take 20 minutes to load
  
• Freeze the system
  
• makes the system sluggish or all of the above when opening the session….
  

Or as I’ll put it: “What exactly is happening with RAM in these larger sessions?”

First, before getting hot under that collar, try a little experiment.  Start a NEW session in your DAW, one that doesn’t have any tracks and isn’t coming from any templates.  Now create an instrument track and instance the ‘problem’ plug-in.  Chances are it loads just fine and now works.  So what is going on?  In a nutshell, you’re session is requiring either more RAM than your system has to offer, or is requiring more RAM than the application can allocate…. huh?

It’s like this…

Currently the most RAM an application ON A MAC can allocate is about 4.5 GB of RAM.   If your system has 4 GB or less of RAM to begin with, then you’ll be hitting that ceiling sooner which is obvious.  What isn’t obvious is that for folks who have 6 or more GB of RAM installed your DAW still can only access a maximum of about 4.5 GB of RAM. So every sampler instrument, Drum kit, orchestral library etc you insert in a session is going to take a small piece of the RAM pie until there is no more and then you’ll get the (read opening sentence)

Well that Sux.

Yes it does, but it’s not a new problem. It’s been with us all along and trust me when I say that today, these limitations seem almost minimal compared to where the technology has come from.  What is new however is the A) every computer can now accept a lot more than 4 BG of RAM, and B) the amount of brand new users who have never used a DAW, never used a Sampler, have never been exposed to the inherent limitations of these systems and have been sold a wonderful “yes you can it’s so easy” dream. Which means when reality hits it’s a shocker for most who really have no idea what’s happening ‘under the hood’

Yes Fine, but what can I do?

Someday in the not very far off future applications will become 64bit. What that basically means, in context of this Blog, is that the limitations of memory access are lifted. Native Instruments for example just released a 64 bit version of Kontakt 3.5 which is 64 bit.  Load as much into it as you have RAM for.  But in the meantime, you’ll need to keep an eye on your overhead.  Don’t load 6 instances of EZDrummer, Don’t load the Acoustic Grand Piano into your templates if you don’t play it. In the case of VSL, they created an application that loads the Vienna library into a second application in the background. This means you get 4.5 GB of RAM just for the orchestra library, and still get to keep 4.5 GB of Ram for whatever DAW you are running.  As long as you’ve got 10 or more GB of RAM to begin with it’s a great solution.

As a final caveat, keep in mind that the 4.5 GB limit I mention here does vary from DAW to DAW, It simply a maximum potential and does not necessarily mean 4.5 GB worth of samples. The DAW itself is taking up RAM and that will vary and some applications may max out at a maximum allocation of 4 or less.

It can be so complicated can’t it?

Get the Flash Player to see this player.

Q: I have a song where we recorded an acoustic guitar first then did the drum tracks later and so on. Before I recorded the drum tracks, I should have fixed the tempo problem first. Can I still fix this problem? (Even after using elastic audio to move some drum hits around?) I don’t work in grid mode or tick base settings for the most part. I know that this needs to change some day. Do you use beat detective?

A: OK, There’s 2 ways I can see going with this.

Way 1) Use Beat Detective to correct specific sections of the song, Say a bad fill for example Beat Detective is good with small sections. and is good for fixing a specific timing problem, but may still not work great if the tempo is all over the place or not defined for the entire song.

Way 2) Go the long way and create a tempo map of the entire song. Allowing you to use Elastic audio on any recorded tracks with more control over changing the actual tempo of the song. This is what I have been doing to fix parts because ultimately everything is much easier to edit in a grid mode once the tempo work is there. Also, Overdubbing new parts might be easier since you’ll be able to use a click track that will correspond to the tempo.

• Perform a SAVE AS at this point, so you can experiment without losing your current place. Creating the Tempo Map involves identifying bars and beats. Start by identifying bar 1. This could be tricky if you have any tracks set to Ticks. You may need to set them back to Samples before Identifying bar 1.
• Next identify Bar 2. This will set the Start tempo of the song. Now create a click track. This will help you by indicating when the tempo is changing and then go through the whole song identifying bars as you go whenever the click track gets off time. Each time you identify a bar, it will adjust the tempo map. Pic 1 shows where the Identify beat command is and also shows you my session, where I’ve gone through and done exactly that. Notice how the Tempo is identified, it’s all over the place. Kicks and snares are good places to identify beat. (Beat Detective can do this too, but I find it very tedious for an entire song)

• 
• Next, Change any Sample based tracks to tick based. Then pick the elastic type for ALL TRACKS - be sure your drum tracks are all grouped. Don’t leave any tracks out because they won’t be adjusted when you change tempo. Then I’ll mess around with the Tempo Operations window.
• In pic 2, I’m selecting 2 bars to apply a tempo change to.

This should go a long way to fixing any tempo issues. I don’t normally quantize things after all this, but often manually adjust or move warp markers using Grid mode. — Hope this helps.

Part 2 of a 4 Part Tutorial covering the Zoom functions in Pro Tools. This Tutorial shows us some Key Commands and covers use of the Zoom Tool.

Get the Flash Player to see this player.

Q: Some friends of mine recorded at a Pro Tools HD studio recently. The studio selected a lower sample rate along w/ using a lot of compression. Why the lower sample rate? Is this common to do?

A: The Short answer is,  to save space and to allow for more DSP processing. The higher sampling rates require twice as much hard drive space, and twice as much CPU processing power.

The Long answer is, it depends.  I’ve run sessions at recording studios  before and I’ve set the Sample Rate to 48k so that when it comes to mixing, I’ll have the ability to apply more plug-in processing. Larger Sample rates may technically sound better, but even with some of today’s best computers, its easy to run out of DSP at the higher sample rates.  Most of the work I do is at 48k.  Deliverables for Television and Motion Picture broadcast are all at 48k. CD’s are all at 44.1k 16 bit.  Mp3s further reduce the amount of info in the audio that is retained.

So the advantages of the higher sample rate for most of today’s work are really limited to the environment its recorded in.  It’s also unclear at this time who can really tell a difference.   There is a much greater jump in quality and dynamics going from 16 bit to 24 bit. That is easy to hear and was a major improvement in digital recording.  But the Jump from 48k to 96k, or even 192k seems to yielded much less of a qualitative improvement.   So to get back to the question, my guess is they felt that 48k or 44.1k would provide them with more DSP for mixing in Pro Tools.  Without talking to the engineer its only a guess.

Another reason is that it takes up half the space on a hard drive. But I do think that for modern rock or electronic music, the lower sampling rate is fine. Just because the rate is lower does not mean quality will suffer. Many many recordings are all done at the lower rates, and the end listener would never know the difference. Good engineering can easily transcend any sample rate.