Sonar Workflow

From Sonar

Revision as of 15:39, 25 September 2006 by Admin (Talk | contribs)

(Back to Home Page)

Contents

Recording Audio

Recording audio in SONAR is quite easy. You simply need to make sure your signal chain is up and working, your sound card control panel is set right, and your track in SONAR is selected as the source and armed.

Signal Chain

This is where it's happening outside the box (e.g. your computer). Simply connect your mic to your mic preamp or mixer, your preamp or mixer to your sound card / audio interface, and connect that to your computer. If you're recording an electric instrument through a direct box of some sort, the process is even simpler: connect your instrument to the direct box IN and your direct box OUT to you sound card / audio interface.

Sound Card Control Panel

Every sound card / audio interface maker has a different control panel applet, but some general rules do apply.

You want to make sure whatever physical IN jack you've connect your mixer or direct box to on your sound card has its volume/input fader all the way up in the control panel, and that it's not muted. Same with the master volume on multi-input sound cards.

Also, if you're using a stereo channel as two discrete mono channels, make sure IN 1 is panned hard left and IN 2 is panned hard right, in both your mixer and your sound card control panel.

Finally, configure your sound card's latency, sample rate, signal levels and other settings in its control panel. Start with a very generous latency, such as 512 or 1024 samples, just to make sure you're not getting instant drop outs in SONAR that keep you from recording.

SONAR track

Select or create a new audio track in SONAR. Go to its input "slot" and select the soundcard and input port your have your equipment plugged into and your sound card control panel configured to use.

Arm the track in SONAR by clicking on the small R on the track's top bar (this is the R that follows M and S). Now hit the record button in SONAR's transport control or the R key on your computer keyboard.

You should now be making music.

SMS 14:02, 15 September 2006 (EDT)

Troubleshooting

Sometimes things just don't work. After all, there are enough cables, boxes, assorted electronic gear and configuration involved in recording audio that anyone can have a mishap. The good news is that once you get your setup working, you shouldn't have to touch it again!

Below is an exhaustive troubleshooting procedure for recording audio, written for users at the "desperate" level. It's based on the Delta44 and Delta1010 audio interfaces, but the suggestions are good for any equipment, though a few specific names of parts may change.

Signal Chain Troubleshooting

Set your signal chain up as follows:

  1. Mic →
  2. Your mixer's Mic IN 1 (make sure mixer is turned on, gain is very high, faders are very high too [just to eliminate low levels as a potential source of problems], and pan is hard left) →
  3. Your mixer's Main Mix 1, Main Out 1, or whatever it may be named →
  4. The Hardware IN 1 on your sound card / audio interface.

Later, you'll also want to connect your mixer's Main Out 2 / Main Mix 2 to the sound card's Hardware IN 2, but that's for later.

TO TEST: Plug your mixer's Main Mix 1 / Main Out 1 directly into a guitar amp or equivalent. If you get sound, you know your mic is working, your cables are good and your mixer is both functional and properly configured. And you can forget about this whole part of your signal chain as a potential source of problems. (You may have to temporarily pan to center to get sound this way).

Sound Card Control Panel Troubleshooting

(The following was written with the M-Audio Delta series in mind, but is easily adapted to other sound cards).

  • Monitor Mixer Page
Make sure that:
  1. All levels are at their absolute highest
  2. Nothing is muted (e.g. everythings is "Solo'd".)
  3. Hardware IN 1 is panned hard left, Hardware IN 2 is panned hard right, and so on, alternating hard left (odd-numbered inputs) with hard right (even-numbered inputs) on both your mixer and the sound card control panel. You can do this any way you please, actually, as long as the mixer and control panel settings are identical.
  • Patchbay / Router Page
  1. Set Hardware OUT 1/2 to "Monitor Mixer" here.
  2. Go to the control panel's "Monitor Mixer" page, stare intently at the colored bars, and sing like a banshee.
  3. Do any of the meter bars move?
  4. If so, your signal chain is working (your cables are good, your mixer is set up right, your sound card is properly installed, you've chosen the right inputs) and your sound card's driver and control panel are properly configured. These, too, can be eliminated as possible problems.

If not, the problem is almost certainly in the control panel configuration. The other alternative is a gear problem -- a loose cable, an short circuit somewhere, etc.

If you've made it this far, go to the next section.

Output Troubleshooting

This is where SONAR comes in. Set it up to use your soundcard as its inputs and outputs:

  1. Go to SONAR's OPTIONS → AUDIO → DRIVERS and select everything that has the name of your soundcard, for both inputs and outputs. Then restart Sonar just to be completely sure the changes have been made.
  2. Go to SONAR's OPTIONS → AUDIO → ADVANCED and make sure the Playback and Recording section's "Driver" item is set to ASIO.
  3. Go to SONAR's OPTIONS → AUDIO → GENERAL and pick the first item listed for your soundcard in both "Playback timing" and "Record Timing". Any will probably do fine, you just don't your onboard sound card or whatever being selected here.
  4. At the very bottom of this page, you will see a button labeled ASIO Panel.... Click it. The M-Audio control panel should appear, and it would be a smart move to double-check all the settings here.

SONAR Troubleshooting

  1. Create a mono audio track.
  2. Set the input to "(Your Sound Card) ASIO" > "LEFT ASIO Analog In 1 (Your Soundcard) [1]" or something very similar to it.
  3. Arm your track: toggle the little R button in the top bar of the track, to the right of the name and the M (Mute) and S (Solo) buttons. It shoudl be red now.
  4. Hit the R key on your computer keyboard or the record button in the SONAR transport controls to start recording. Sing away.
  5. Stop, go back to the beginning, and SEE if anything was recorded -- look at the wave form in the track and if it's not flat, you (or your audio, to be more precise) are getting in! (Make sure OPTIONS → GLOBAL OPTIONS → GENERAL → "Display Waveform Preview While Recording" is checked to see things in real time).

If you see your sound here, everything from your mic on through your sound card and Sonar's INs are properly configured. Breathe a sigh of relief. But don't break out the six-pack yet.

Now, if you can also HEAR your audio upon playback, your problems are solved. But assuming you can't:

  1. Go to your audio track in SONAR.
  2. Go to the "Output" section (the horizontal bar with a stylized O at the far left).
  3. Click it and select Master.

You could go out directly here, instead of throught the Master, but that will probably only cause you problems later. So we're going to do it right, though it takes another step or two. So, you've selected Master. Now...

  1. Go to the bottom of the SONAR screen and you should see a buss (looks just like a track really) that says Master.
  2. On ITS output select "ASIO Analog OUT 1/2 (Your Sound Card) [1]" or similar.
  3. Plug your left studio monitor's audio cable into it at one end and into HW OUT 1 of your sound card at the other.
  4. Plug your right studio monitor's audio cable into it at one end and HW OUT 2 of your sound card at the other.
  5. Plug your monitors in and turn them on, if you have active (powered) monitors. Do the same with your monitor amplifier if you have passive (unpowered) monitors.
  6. Play your recording, cranking the volume as needed.

You should hear something here. If you've followed all the above steps, AND you don't happen to have faulty hardware, it's inevitable.

Assuming you are getting sound, now you'll want to use different outs -- at least one direct to your monitors (you've just done that) and one to your mixer for the headphones used in tracking.

That's easy! Just...

  1. Go to SONAR's Master buss.
  2. Right click on some grey part of it.
  3. On the pop-up menu, select Insert Send.
  4. Follow that menu to "(Your Sound Card) ASIO Analog Out 3/4 [1]" and select it.
  5. Now connect your sound card's Hardware OUT 3 and Hardware OUT 4 to the appropriate input on your mixer.

Wasn't that a great cop-out -- "appropriate input"!

  1. On smaller mixers the TAPE/CD IN L & R inputs are commonly used. This often requires a custom cable using two 1/4" TRS cables at the soundcard OUTs end and two RCA jacks soldered on to the other end and plugged into the TAPE/CD IN L & R.
  2. On many mixers this also requires the Main / CD-Tape button be be down in order to work.
  3. You may have another IN on your mixer that works for this. If so, no soldering or RCA jacks will be needed -- just plug the TRS cable into the mixer.
  4. Plug headphones into mixer's Phone Out plug (or something similarly named).

If you follow these directions to the letter you SHOULD be good to go. Otherwise, you may have a problem with your gear.

Gear Troubleshooting

In the first case, you will have to test every single physical component of your system. That means either swapping each component out for an identical one, or testing each component on another system:

  1. Mic / instrument
  2. Every single bloody cable you have
  3. Mixer / direct box
  4. Sound card breakout box (if applicable)
  5. Sound-card-to-computer cable
  6. The sound card itself

When you do this:

  1. Test one component at a time
  2. Make very sure that everything is plugged in completely and solidly.
  3. Move, jiggle and shake each one to make sure there's no short anywhere (who knows -- maybe your out jacks' soldering is coming loose or something along those lines).
  4. Use whatever type of cables your manufacturer suggests. Using 1/4" unbalanced (TS or guitar) cables when 1/4" TRS (balanced) cables are specified might work, but it also might not. Don't take the chance -- use what the manufacturer suggests. Using balanced cables will also drastically reduce the noise in your recordings, so it's double plus good!

TIP: When troubleshooting anything, you must be extremely disciplined and systematic. Change one thing at a time (no more!) and then test your entire rig out. Then change another and test again. And so on. It's tempting to change a bunch of things at once, in hopes of saving time, but though you may get lucky this way, chances are you won't, and you also won't have any more of an idea of what's wrong than when you started.

SMS 13:58, 15 September 2006 (EDT)

Recording MIDI

From Softsynths, Samplers, Etc.

From External Controllers / Synths

Mixing Procedures

Exporting Procedures

Exporting From Sonar

Once you've finished composing, tracking, mixing and exporting your music, you'll end up with one of four file types with SONAR's File | Export | Audio function: RIFF WAV, Broadcast WAV, Windows Media Advanced Streaming or MP3 (if you have paid separately to use the MP3 codec that comes with SONAR).

RIFF WAV and Broadcast WAV files are the file types of choice for mastering, as they are lossless formats.

MP3, on the other hand, is a lossy format, meaning that you not only lose information (read: quality) by encoding a file as MP3, but you also lose additional quality every time you modify the file's content (though not, of course, its metadata). This makes MP3s suitable for casual listening, but not for mastering or even demanding listening.

If you choose not to buy the shareware codec that comes with Sonar, you can still produce MP3s with an external codec. By far the most highly recommended of these is LAME, which is open source. It is also widely considered to be the best MP3 encoder around, making it preferable to the codec bundled with SONAR even without condidering monetary factors.

SMS 14:02, 15 September 2006 (EDT)

Exporting MP3s from SONAR using LAME

Step-By-Step Instructions For Encoding With LAME in SONAR

If you don't want to unlock (buy) the Cakewalk MP3 encoder, you can use the LAME MP3 encoder directly in SONAR. This way you don't need to first export to .wav and then encode to .MP3. The LAME MP3 encoder is high quality, open source and free.

Here are step-by-step instructions for setting up the LAME MP3 encoder (from the Cakewalk Forum Thread by Jesse G):

  1. Download and install the LAME encoder.
  2. Open SONAR 3/4/5 and go to Tools → Cakewalk Ext. Encoder Config.
  3. In the Friendly Name field write LAME MP3 Encoder.
  4. In the Extension field write .MP3.
  5. In the Description field write whatever you want.
  6. In the Path field put the path where you installed the LAME encoder. That is where lame.exe and lame_enc.dll are located. I keep my LAME encoder in my "C:\\Program Files\\Cakewalk\\Shared Utilities\\LAME" folder. You can also copy and paste the extracted contents of the LAME .zip file to your Sonar shared utilities folder and then, when you have to perform this step, click the browse button to find the location.
    (Note that you may have to enclose paths containing spaces in quotation marks).
  7. In the Command Line field write (for example): lame -b 192 -m j %I %O (See below for more encoding options).
  8. Uncheck Keep Wave File.
  9. Press Save button.

This will export a constant bitrate 192Kbps joint stereo MP3 file. If you want 128Kbps, replace 192 with 128 in the Command Line field. The next time you click on File → Export → Audio in SONAR 3, 4, 5 you'll find "LAME MP3 encoder" as a type of audio.

If you want to be able to export to both 192Kpbs and 128Kbps, click New and then repeat the above steps, but this time use 128 instead of 192 in the Command Line, and also give them different Friendly Names, such as "LAME 192k MP3" for the 192 version and "LAME 128k MP3" for the 128 version. Now you will have both choices the next time you export audio.

Dcastle 12:42, 15 September 2006 (EDT)

Fine-Tuning LAME Encoding

LAME permits a huge number of options in the command line. It's well worth your while to check out the main ones, as they can greatly affect the quality and file size of the MP3s you produce. These options are specified in the Command Line field of SONAR, as mentioned above. The syntax is:

lame <options> <infile> <outfile>

In practice, you will only be interested in changing the options, so the syntax can be simplified to:

lame <options> %I %O

LAME offers three types of encoding: Variable Bitrate (VBR), Average Bitrate (ABR) and Constant Bitrate (CBR). The rule of thumb, following Hydrogen Audio's testing, is as follows: at a given bitrate, VBR is higher quality than ABR, which is higher quality than CBR (VBR > ABR > CBR in terms of quality). The exception to this is when you choose the highest possible CBR bitrate, which is 320 kbps (-b 320 = --alt-preset insane), but this produces very large file sizes for very little audible benefit.

The following are the recommended LAME encoding settings from Hydrogen Audio. They all go in the <options> field of the Command Line entry in SONAR:

Variable Bitrate (VBR)

Use VBR when the goal is to achieve a fixed level of quality using the lowest possible bitrate.

  • VBR is best used to target a specific quality level, instead of a specific bitrate. The final file size of a VBR encode is less predictable than with ABR, but the quality is usually better.
  • Usage: -V(number) where the number is 0-9, 0 being highest quality, 9 being the lowest.

Example: lame -V2 %I %O

Average Bitrate (ABR)

ABR is a compromise between VBR and CBR modes. ABR encoding varies bits around a specified target bitrate.

  • Use ABR when you need to know the final size of the file but still want to allow the encoder some flexibility to decide which passages need more bits.
  • Usage: --preset <bitrate> where <bitrate> (desired averaged bitrate in kbit/s) can be any value between 8 and 320 (e.g. 9, 17, 80, 128, 133, 200, etc).

Example: lame --preset 200 %I %O

Constant Bitrate (CBR)

CBR encoding is not efficient. Whereas VBR and ABR modes can supply more bits to complex music passages and save bits on simpler ones, CBR encodes every frame at the same bitrate.

  • CBR is only recommended for usage in streaming situations where the upper bitrate must be strictly enforced.
  • Usage: -b <bitrate> where <bitrate> (in kbit/s) must be chosen from the following values: 8, 16, 24, 32, 40, 48, 64, 80, 96, 112, 128, 160, 192, 224, 256, or 320.

Example: lame -b 192 %I %O

Fine-Tuning Other LAME Options

When exporting MP3s from SONAR using the LAME encoder, you can have a series of convenient operations performed automatically. These include:

  • -c: Setting the Copyright bit.
  • --clipdetect: Detecting clipping in the encoded MP3 file (you will only see this information if you keep the LAME command-line interface open).

Example: lame -V2 --clipdetect -c %I %O

This will produce a very high quality VBR (-V2) MP3, with the copyright bit set and clip detection activated.

The LAME help files detail many more options.

Getting LAME

The main LAME site, on SourceForge, offers only the uncompiled LAME source code, which is of no use to the average end user. What you need is a compiled binary of the LAME encoder.

You can download the LAME encoder from several sites, including:

Avoid using alpha (a) versions of LAME. More often than not these are exclusively for testing purposes -- use them only if you want to help developers with feedback.

To manually encode MP3 files using LAME, several graphical front-ends are available, including RazorLame. Note that RazorLame and most other front-ends require that you install the actual LAME binary encoder separately.

Exporting Lossless FLAC files from SONAR

Step-By-Step Instructions For Encoding With FLAC in SONAR

To export compressed archival-quality files from SONAR, the FLAC encoder is recommended. Lossless formats such as FLAC retain all the information contained in the source .WAV --the FLAC file can be converted back into a .WAV file with 100% bit-for-bit accuracy-- but are highly compressed, though not nearly as much as MP3 files.

Here are step-by-step instructions for setting up the FLAC encoder. They have been tested in SONAR 5.2 Producer Edition.

  1. Download and install the FLAC encoder.
  2. Open SONAR and go to Tools → Cakewalk Ext. Encoder Config.
  3. In the Friendly Name field write FLAC Lossless Encoder.
  4. In the Extension field write .fla.
  5. In the Description field write FLAC Lossless Compression Level 5.
  6. In the Path field put the path where you installed the FLAC encoder. This is where flac.exe and related files are located.
    (A path without spaces is recommended).
  7. In the Command Line field write: flac.exe --compression-level-5 %I -o %O
  8. Uncheck Keep Wave File.
  9. Press Save button.

This will export an optimally-compressed (compression level 5) FLAC file with no frills.

If you want FLAC to automatically add the few tags that SONAR passes to external encoders, use these options in Step 7:

  • flac.exe --compression-level-5 -T "ARTIST=%A" -T "TITLE=%T" -T "COPYRIGHT=%C" -T "COMMENT=%D" %I -o %O

This puts the information found in SONAR's File → Info dialog box into the appropriate tags of the FLAC file (with "Description" being placed in the Comment field for lack of a better place).

In addition to the above tags, which will be different for each tune you compose (assuming you actually update the SONAR Info box), you can add your own tags with static data:

  • flac.exe --compression-level-5 -T "DATE=2006" -T "ALBUM=My Rockin Sonar Tunes" -T "GENRE=Cliched Heavy Metal Power Ballad" -T "ARTIST=%A" -T "TITLE=%T" -T "COPYRIGHT=%C" -T "COMMENT=%D" %I -o %O

Here, you've set a fixed date with -T "DATE=2006" (which will have to be updated annually, of course), a fixed album name with -T "ALBUM=My Rockin Sonar Tunes" (which you will want to update as you move from one record to another) and a fixed genre with -T "GENRE=Cliched Heavy Metal Power Ballad" (which will require profound personal evolution and deep self-examination in order to change). Plus, all the automatic tags will be there, too.

NOTE: There is a limit to the number of characters that SONAR can pass to a command-line encoder. If you get an error when trying to encode using the tag settings above, you must either remove some of the tags from the command line, or reduce the length of the text they contain (e.g. a 12 paragraph description in the SONAR Info box would cause the encoding process to fail).

Fine-Tuning FLAC Encoding

FLAC permits a huge number of options in the command line, but the truth is you won't really need them. However, in case you decide to experiment, read the FLAC documentation for more details.

Although FLAC offers 9 levels of compression (0-8, with 0 being the lowest compression and 8 being the highest; audio quality is identical in all cases), the optimal level is 5. Higher than that and you gain only a small increase in compression, at a large processing cost. Lower than that and the level of compression quickly becomes less than ideal. But at the same time, since FLAC is an asymmetric encoder, encoding a FLAC file at higher compression rates takes up more CPU time, but decoding one (e.g. listening to it) does not take appreciably more CPU resources.

Getting FLAC

The main FLAC site, on SourceForge, has everything you need to use FLAC -- the encoder binaries, documentation, a FAQ, and so on.

Listening to FLAC

FLAC is admittedly not the world's most common format for listening, but it's quite easy to do so.

  • Foobar2000 is a free audio player that supports FLAC and a ton of other formats you probably haven't heard of, right out of the box.
  • VLC is a GPU video and audio player that supports FLAC out of the box.
  • Winamp can play FLAC files by installing a simple plug-in.
  • Windows Media Player can play FLAC files by installing either the CoreFlac of Illuminable DirectShow filters.
  • J.River Media Center can play FLAC files using the In_FLAC plug-in.

There are also FLAC plug-ins for many other media players. See here or here (at the bottom of the page) for lists of programs and plug-in links.

Transcoding Into Other Formats

You may want to transcode a file exported from SONAR into another format for any number of reasons -- getting better sound per byte than MP3, getting the same quality as MP3s in a smaller size, archiving mixdowns, and so on. To do this, you need an external codec, either lossy or lossless. These may have built-in graphic interfaces for easy transcoding (drag-and-drop, for example), or they may be command line utilities, with or without a graphical front-end.

Lossy Formats

You can convert a (RIFF) WAV file exported from SONAR using any number of lossy converters.

These include OGG/Vorbis, AAC, Dolby AC3 and Musepack.

WMA gives you the choice between lossy or lossless compression; the lossy compression is roughly comparable to that of MP3 files at equivalent bitrates, and the same caveats for the MP3 format apply to WMA lossy compression.

Comparisons of lossy codecs are available here, here and here.

Lossless Formats

Assuming you wish to keep archival (lossless) copies of your mixdowns, but want to use a minimum of space, you have the option of converting RIFF WAV files to a compressed lossless format (as well as exporting directly), which can provide a considerably smaller file than the source WAV, while retaining 100% of the audio information in the recording, unlike MP3 or WAV lossless files. FLAC also provides file integrity information through frame CRCs and MD5 signatures.

One of the most popular lossless encoders is the open source FLAC. FLAC is available for most every operating system around, and in addition to providing substantial lossless compression, it is also directly playable on a growing number of home and car stereos, portable music players, and software music players. Furthermore, it is streamable and seekable. Some find it to be the ideal compromise solution for digitizing their CD collection, providing perfect playback while taking up less space than WAVs (though quite a bit more than MP3s).

FLAC files can be converted back into WAV files that are bit-for-bit identical to the source WAV using the command line encoder (which is also a decoder, tester and analyzer), various front ends, or the built-in file converter functions of some software players, such as Foobar2000.

The FLAC command-line encoder can be downloaded from the FLAC homepage. A large number of graphical front-ends for making FLAC easier to use are available here.

A filter that allows FLAC files to be edited directly in Cool Edit or Adobe Audition can be had here. See above for more details on playing FLAC files (not necessary for archival purposes, but it's pleasing indeed to hear true CD quality sound in a compressed format).

Other lossless codecs are Monkey's Audio, WavPack and WMA Lossless.

Comparisons of FLAC and other lossless codecs can be found here, here, here and here.

SMS 14:02, 15 September 2006 (EDT)

Long-Term Archiving and Storage of Projects

Formats

CWP + Audio Files Folder

It is highly recommended that you select "Use Per-Project Audio Folders" in SONAR's Options → Global → Audio Data page (it's at the very bottom, under Per-Project Audio). This puts all the audio files for a given project into their own directory (er, folder), with the project file (.cwp) in the project folder directly above the audio folder. In other words, this option creates a directory structure like the following:

  • E:\\MySonarProjects
    • Killer_Song_01
      • Audio
        • sound001.wav
        • sound002.wav
        • sound003.wav
        • sound004.wav
      • Killer_Song_01.cwp
    • Rocking_Dog
      • Audio
        • sound001.wav
        • sound002.wav
        • sound003.wav
        • sound004.wav
      • Rocking_Dog.cwp

As can be seen from this, per-project audio folders make it quite easy to back up projects -- you just copy the top-level folder of your songs (the Killer_Song_01 and Rocking_Dog folders in this case) somewhere else.

The alternative to per-project audio files is having all the audio files from all your songs put in one massive folder, with file names that give you absolutely no clue as to what song they belong to. There is absolutely no advantage to the "all audio in one folder" approach.

Note that when you back up a project while using per-project audio folders, the audio data is not compressed in any way -- it's in simple .wav" files. This can make for a massive amount of data. To get around this, you can use ZIP, RAR or some other type of compression to drastically reduce file sizes. There is a serious caveat to this, however -- the omnipresent danger of putting everything in one big file, which means that a single corrupted byte can leave you with absolutely nothing from your project.

To get around this there are two solutions. First, you can use an applications such as WinRAR which allow you to compress each file to a single archive. You'll end up with the same number of files in the same places this way, but any corrpution will be partial -- you may lose a file or two to a dying DVD, but not your entire project.

The other solution is to create parity files using QuickPAR, which will allow you to recreate with perfect accuracy entire files or even corrupted parts of a single huge file.

CWB/BUN Bundle Files

Whether you use per-project audio folders or not, you can save all the audio files, as well as the project file, from a given song in one large "Bundle" file, which will have the extension .bun or .cwb.

This approach has the advantage of putting everything you need for a given song (except external programs, such as softsynths and plug-ins) in one single place -- one single file, to be exact. However, a very large number of SONAR users are vehemently opposed to this solution, as bundle files are known for becoming corrupted, and of course face all the problems that come with storing something as one big file.

To get around the dangers of having one huge file, QuickPAR is strongly recommended. As to corruption caused by SONAR itself, either when creating or opening bundle files, this is an open question, and there is no known solution for corruption caused this way.

Therefore, the CWP + Audio Files Folder backup method is highly recommended.

Mixdowns

Along with your SONAR project files, which allow you to continue tracking, mixing, twiddling, and so on, it is a good idea to create and store a mixdown of your project in its current state. This way you will always have some version of your song in complete form, since while your project files may work in every future verison of SONAR ever made, your plug-ins and virtual instruments may become lost, you may forget to install them, or they may become incompatible with future operating systems.

At the very least, having a mixdown will allow you to hear what you created so many years ago and approximate the same sound (if that's what you're after) with different plug-ins and virtual instruments.

And again, to reduce file size you can compress the mixdown file as a ZIP or RAR archive, or you can export it directly as a lossless format such as a FLAC file for even better compression.

Physical Media

Once you've decided on a format to back up to, and after you've hopefully created several hundred MB of parity files with QuickPar, you still have to find somewhere to put all these files. This is a far more complicated issue than many believe, as no storage medium is perfect -- all will fail eventually.

This bears repeating: every single storage medium in existence will, sooner or later, fail.

With that in mind, there are ways to can minimize your risk, the easiest being to make multiple backups (hopfully on different types of media) and store them in different places. If you make 5 backups on bad DVDs, they will all most likely die at more or less the same time. And if you have 5 backups on 5 different types of media that are all stored in your house, a fire will quickly eliminate them all.

Hard Drives

Hard drives are fast, cheap and can hold a vast amount of data. However, when they die they often do so catastrophically, meaning quickly and violently. The upside of this is that files can often be recovered from a crashed hard drive with realtive ease -- if you have several hundred or thousand dollars to pay a professional data recovery service.

Hard drives' tremendous convenience often outweighs these considerations, though, and they are often the go-to backup medium for DAW users.

Onto Two Drives At Once

This is not strictly a backup method, but a way of reducing your chances of losing your current work. It involves using two or more hard drives simultaenously to store the same data, and it is known as RAID. With a mirrored RAID array, if one of your drives dies, the other one keeps on ticking and you can keep on working as if nothing had happened. Then, when you can (hopefully soon after the drive dies) you remove the dead hard drive and replace it with another one of equal or greater capacity, and the RAID software (or hardware, depending on your setup) copies the data from the good drive to the empty new one, and your safe again.

Onto The Same Drive Your Project Is On

To save a mix you are particularly fond of before going for that extra level of mixing perfection, for example, backing up onto the same hard drive you use for your SONAR projects, or a different partition on that same drive, may be a practical option. This is risky -- you will lose everything if the drive fails, and the same may happen if your computer is the victim of a virus, an idiot who knows what "format c:\\" does, a nasty power surge, or physical damage.

Think of this as an makeshift version control system more than a backup solution.

Onto A Different Drive In The Same Computer

Backing up your project onto a different hard drive than your project files are on is a (wee) step up from using the same drive for this purpose. It eliminates hard disk failure as a potential cause of lost backups for all intents and purposes (the odds of two drives failing simultaneously are quite low).

But this method still leaves you vulnerable to viruses, idiots, power surges, physical damage -- in short, anything that can damage your audio project drive can damage your backup drive if it's in the same computer, except simple, spontaneous single-drive failure.

Again, this is more of a kludged versioning system than a true backup solution.

Onto A Different Drive In A Different Computer

Now we're starting to enter into genuine backup territory. Putting your backup files on a different hard drive greatly decreases the chances of something happening to both your working and backup files at the same time.

Unless the second hard drive is physically connected (through a network) or on the same part of the power grid (e.g. your kid's computer, in your house, even if it is not networked with your DAW). In these cases power surges can very well affect both drives at once, and you're up the proverbial creek again. Nasty viruses that spread over networks can be another cause of disaster (and you know your kid's computer is chock full of spyware, trojans, viruses, evil macros, malicious toolbar components and other malware).

Onto An External Hard Drive

This is by far the safest hard drive-based backup method. You store your backups on one or more external hard drives (typically Firewire or USB, though even IDE can be used for this).

With an external hard drive, you plug the drive in, copy over the files to be backed up, unplug the drive, and keep it somewhere safe -- hopefully off-premeses. Do this with more than one drive and your chances of not losing everything at once increase tremendously.

For long-term storage (archiving), it is recommended to plug your drive in and spin it up anywhere from one to four times a year, to prevent stiction.

DVDs

With the advent of cheap DVD burners and media, many thought that the backup issue had finally been solved by this high-capacity medium -- burn your files onto a DVD and forget about them.

Unfortunately, DVDs are extremely prone to partial or total failure, sometimes after as little as two months. Of course, some can last for many years. The problem is you never know which are the good ones and which are the bad ones.

Some general guidelines for using DVDs as archival media:

Always Back Up To Two Or More DVDs

What's it cost? Maybe 25 cents more, plus 5 minutes of your time? Do it!

Always Record Several Hundred MB of Parity Files Along With Your Project Backups Using QuickPar

This will allow you to recover corrupted files from dying DVDs.

Use a real DVD Tester

The "Verify Disk" functions of many (if not all) DVD burning programs are absolutely and totally useless for verifying burn quality. They simply verify that a disk is readable -- and readable lumps utterly perfect DVDs together with ones that are one mote of dust away from total failure, along with everything in between. A DVD declared "readable" by such a program may fail the next week.

To remedy this, verify every recorded DVD with a true DVD tester, such as Nero CD-DVD Speed's Disc Quality function. Such programs read every sector on the DVD and tell you exactly how many errors there are in each, using a nice and very readable graph. Fewer than 280 PIE errors and 4 PIF errors per sector are considered a good result.

Handle Your DVDs Properly

DVDs are even more sensitive to mishandling than CDs, as well as most newborns. Treat them right!

  • Do not put labels on DVDs. The adhesive can damage the written layer (which, contrary to popular belief, is right below the TOP face of the DVD, the opposite side from which it is written) and anything less than perfect positioning can make the DVD wobble.
  • Do not write anywhere on a DVD except the clear plastic inner ring. No data is stored there. And only use a special felt-tip marker for this.
  • Never touch the top or bottom surfaces of a DVD. Scratch the bottom and the laser has trouble reading the data. Scratch the top and you can easily destroy the data physically.
  • Store DVDs in a dark place. UV rays can damage them.
  • Store DVDs in a dry place. Mold can grow on them in damp places, adhesives can begin to deteriorate, and so on.
On these last two topics, NIST says: A temperature of 18°C and 40% RH would be considered suitable for long-term storage. A lower temperature and RH is recommended for extended-term storage.
  • Store DVDs upright.
  • If you have to clean a DVD, do it with a lint-free cloth using distilled water, isopropyl alcohol or methanol. Clean from the center hole outwards, never in a circle.
Reverify Your DVDs Periodically

Use the same real DVD testing program periodically to make sure your disks are not deteriorating. If they are, recover your data and re-backup as soon as possible. Doing this every three to six months would not be unreasonable.

Use The Latest Firmware For Your Burner

Unlike firmware for many other devices, DVD burner firmware is almost always benign -- the chances of your frying your drive are next to nil.

DVD firmware consists basically of improved write strategies for different media, and/or entirely new write strategies for new and old media alike. Often, these write strategies also allow faster burning of existing media.

Newer firmware almost always improves burn quality, at least of the types of media each update covers.

Use Quality Media

Logical. Makes sense -- why save a few dollars when the ultimate price may be your demo or album? The problem is that it's impossible to identify quality media in the store, or at home for that matter, unless you go to almost ridiculous extremes.

It's a dirty little secret in the industry that the DVDs you buy from Memorex, Fuji, Imation, and any other brand are almost never made by them. These big brand names, as well as all the small fries, are basically mere resellers, having their labels slapped on DVDs made by other companies. In other words, there is no such thing as a good brand of DVD -- only good manufacturers.

Ergo, there is absolutely no correlation between DVD brand and DVD quality.

Further complicating matters, a given brand (the names you trust!) will typically sell DVDs made by multiple manufacturers -- for example, "Memorex" 8x DVD+Rs sold in packs of 10 jewel boxes, in 25-packs, in 50-packs and in 100-packs may well all be made by different companies, and may have vastly different levels of quality. And you have no way of knowing.

Actually, there is one way. To find out the true manufacturer of a given DVD must use special software that reads DVDs Media ID code, which is stored on every disc. Only then will you know who made your DVD -- and you'll probably be surprised. Ever heard of Daxon, Moser Baer, Interaxia, Nanya Tech, CMC Magnetics, LeadData or Mitsubishi-Kagei? These are some of the real DVD makers.

But keep in mind that more than a few manufacturers fake the Media ID code on their DVDs, putting the code of a more prestigious maker onto their own junk.

To find quality DVD media you have to buy blind, test like crazy, and when you find some that works well you have to run back to the same store and buy a couple hundred dollars worth of the exact same brand of the exact same speed of the exact same type DVD, making sure what you buy has the exact same packaging and format as what you tested (the 100-pack is probably made by a different company than the 10-pack). Oh, and you have to hope it's all from the same lot, because resellers change suppliers all the time.

DigitalFAQ has an excellent guide to who makes whose DVDs, as well as links to software for identifying DVD manufacturers. Not that this is any guarantee, as they themselves would be the first to point out.

The same site also has a very good series of FAQs and guides to other DVD-related issues.

Finally, hardware matters. Some DVDs that burn perfectly in one DVD recorder die a quick death when burned in another. So when you do find that killer media after extensive testing, don't go and tell your friends about the great DVDs you found -- they may well be mediocre or even garbage when put in your friends' burners.


If you want to delve very, very deep into this issue, the US National Institute for Standards and Technology (NIST) has a 50 page study called Care and Handling Guide for the Preservation of CDs and DVDs available for your reading pleasure.

Burn at the Right Speed

Burn your DVDs at the right speed. But what is the right speed? The only way to know is by testing each lot of DVDs on your own computer, unfortunately.

The combination of the burner you use, its current firmware, and the DVDs you're writing to all affect the right write speed. In some cases, burning 8x media at 8x media produces optimal results, and in others it produces a disaster as the dye dies due to the high laser power used to write at fast speeds. In other cases, writing to an 8x DVD at 1x may also produce a dead or dying disc, possibly because the write strategy for that particular media in your particular burner's current firmware is not very dependable.

So again, it comes down to systematic trial and error. Recomendations like "burn at 1x" or "burn at half the media's rated speed" are simply meaningless.

Tape

Tape?

Online Services

Storing your music online is not a common alternative, for several reasons: security, speed and size.

You obviously don't want your demo or latest hit to fall into the wrong hands, and you can never really be sure what happens to them once stored at an online service. Such services may also be hacked. To keep your music from getting into the evil clutches of... whoever... you can encrypt the files first, using PGP or other software.

Upload speed is also a problem, and it is exacerbated by the large file sizes involved in digital audio. But if you do decide to back your projects up online, PC World, PC Mag, Yahoo and Lights.com all have lists of free and paid online storage services.

Archiving Security Software

Whatever format and physical media you choose, there are additional steps you can take to protect your valuable (or invaluable) recordings.

  • QuickPar
QuickPar is a free utility that scans your files, divides them up into virtual blocks, and creates a series of PAR (parity) files that allow you to recover corrupted files from any type of media. It works on the same principles as RAID.
For example, if you choose to use virtual blocks 1MB in size, create 500MB of QuickPar files, and burn them along with 3.8GB of SONAR projects onto a DVD, you can recover up to 500 corrupted file chunks of up to 1 MB in size (it doesn't matter if the chunk has 1 byte or 1 MB of corruption). In other words, you can save up to 500MB of audio files that have been utterly corrupted by a dying DVD -- files you would never be able to access again otherwise.
This is very, very good insurance. And it works for any type of physical media.
  • SFV
SFV, or Simple File Verification, scans your files and creates a checksum for them. This allows you to determine if any of them have been corrupted. But unfortunately, it doesn't have the capability to recover corrupt files. For that, use QuickPar.
Lists of SFV utilities for download are available here and here. Most are freeware.
  • MD5
MD5 does the same thing as SFV, and has the same limitations -- it can tell you your files are corrupt, but it can't do anything about it. Again, for that use QuickPar.
A free MD5 utility is available here.

SMS 14:02, 15 September 2006 (EDT)

(Back to Home Page)













Retrieved from "http://editthis.info/sonar/Sonar_Workflow"
Personal tools