What goes into creating a comprehensive expansion pack? It’s not as easy as it might seem…

If only I’d known what I was getting into…

When I first played the EB 5-String Bass, I loved the sound and wanted to sample it for my own use. In the process, I created a Dimension Pro instrument and made it available for free to the Cakewalk community as a “thank you” for all the support you’ve given SONAR.

But also in the process, I found out the EB uses a unique Tuned Coil Tap technology for the two pickups, yielding a total of eight distinct sounds. I found them all useful, so of course I wanted to sample those before the loaner bass went back.

And that’s how the new Gibson EB-5 String Bass Expansion Pack started. If you’re interested in what goes into creating a Dimension Pro instrument…read on.


First was the decision of what/how to sample. Individual notes? Fewer notes and transposition? Looped? Full sustain? Ultimately, I decided to go for the purest sound possible, which meant sampling every single note individually (41 total; the EB 5-String goes to B below a conventional 4-string bass) and letting them decay naturally—no looping. Also I didn’t synthesize note pitches, even though I found that the iZotope Solo Bass algorithm could do so without creating the annoying format shifts you usually get with sample transposition. That’s all well and good, and I used that technique on the free instrument to generate artificial notes below B, but I wanted each note to have its own character.

The other decision was resolution, and I went for 24-bit/44.1kHz. With bass, there’s certainly no reason to have a super-extended frequency response; aside from being an extremely common sample rate, 44.1kHz was more than adequate.


I did a “bass sampling setup”—super-high bridge to prevent any string rattling. That didn’t make it easy to play the notes, and of course, required an intonation adjustment to compensate. I went direct into the V-Studio 700 aux input, maintained a decent amount of headroom, and played each note in sequence on a single track. TH2 Producer was in the FX bin and open at all times so I could verify that the notes were in tune.

Sampling was not just about hitting each note. If there was any buzz, I did the note over. I kept my eyes glued to the screen to make sure the levels were relatively consistent, and was very careful to find the optimum picking place on the string for each sound. (Usually this was between the neck and bridge pickups, but some sounds sounded better when picked closer to one pickup or the other.) A thick thumbpick gave a strong attack, because I wasn’t planning on sampling at different velocities. (I find it easier, and more predictable, to record with a strong attack and then make it less prominent for softer velocities using Dimension Pro’s editing options—mostly pulling down the 1-pole lowpass a bit, and kicking it up with velocity.) Many notes required multiple plucks (sometimes a couple dozen) to get one that was “just right,” at which point I’d proceed to the next note.

One aspect I didn’t expect was the insane amount of sustain. It’s not just the wood; the EB basses have a Babicz bridge, which is massive. Many notes sustained longer than 75 seconds (I lost patience waiting to find out what happened after the string level hit -66dB), so realizing that it would take an excessive amount of memory and that notes of that length had limited musical usefulness, I sampled each string until it reached what I thought was enough—typically around 15 seconds.

While I was at it, I also recorded down slides and up/down slides for each string, as well as one “effect” sound. These were all different for the different instruments, and I allowed myself some fun with the effects—harmonics, hitting the strings, muted sounds, whatever.


The Tuned Coil Tap settings retain some, but not all, humbucking properties so the bass would pick up some extremely low level computer noise and such under these conditions. After I’d finished recording the file with all the notes, I ran it through Sound Forge’s noise reduction algorithm which essentially turned the interference from almost inaudible to non-existent.

Interestingly, I didn’t have any interference when both pickups were set to the Tuned Coil Tap position. Curious, I talked to Gibson pickup guru Jim DeCola, and he explained that the two pickups are reverse wound and reverse polarity; when used together, the pair of pickups basically act like a giant humbucker. Well, I can confirm it works.


To make my life easier (remember…eight instruments, each with 41 notes!), I set up a template in SONAR with a track for each note (Fig. 1).

Fig. 1: The bass sampling template greatly simplified the process of sorting and editing the samples for the various notes.

I brought the de-noised file back into SONAR, then cut the individual notes from the file and placed them in their appropriate tracks. Colorized tracks were very helpful here, as the notes from each string had their own color.

Then came what was arguably the most tedious part of the process: trimming the beginning of every note about 5ms before the note began, then adding a 5ms attack time to insure there wouldn’t be any clicks from a level offset. After the attacks were trimmed, then it was time to trim the ends and create decays. Again, this was extremely tedious as I listened to every note, and had to decide where the decay had become so low in level the note didn’t need to continue past that point.

Once the notes were all trimmed, I then dragged them into a folder that the SFZ file could reference. The template had the note name for each track, so the file names contained that but they also carried over the project name so I had to rename all the files to contain only the file names the SFZ file would reference…more tedium, but it needed to be done.


The only real effort here was creating the first SFZ file (Fig. 2), as I could use it as a template for all the subsequent instruments.

Fig. 2: Part of the SFZ instrument definition file for one of the bass instruments.

I made two main tweaks other than simply assigning notes to keys and referencing folders : adding a short release time to avoid any key-up clicks, and reducing the level of slides so they’d blend better with the notes when playing parts.


I have a tendency to go nuts with programs, so I made a decision to create a basic, pure program with no significant gimmicks. As with the SFZ file, it was possible to make a basic program and just adapt it for the different instruments. Aside from using the filter to give more of a sense of dynamics, I used keymapping on the Amp to reduce note level as you play higher up on the neck, and also tied the Mod Wheel with EQ1 and EQ2 to emulate the effect of pulling back the EB tone control.

This worked by pulling down the EQ1 high-shelf gain while simultaneously adding a slight boost with the EQ2 bandpass filter to emulate the resonant peak that occurs when the tone control cap forms a resonant circuit with the pickup (if you’re curious about how tone controls really work—it’s not just about rolling off highs!—check out this article I wrote for the Gibson.com site).

However I just couldn’t resist adding another set of programs that engaged the Drive to give a variation with a bit more “growl.” Sure, you could always run Dimension Pro through an amp sim, but the aggressive, more “amp-like” sound is very useful in arrangements and it’s convenient to have the second set of “growl” patches (Fig. 3).

Fig. 3: The Dimension Pro instrument, with the Browser open to the collection of programs from the EB 5-String Bass Expansion Pack, as well as the instrument program available as a free download.

There was one more issue, though (nothing’s ever easy, right?). I QCed the pack by playing every note of every instrument, and found that some gave a nasty, digital distortion spike. After looking at them in Sound Forge, I saw that they hit 0. They didn’t appear (nor sound) clipped; so I just reduced the gain by 0.1dB, and the problem went away.


Now all that was left was writing the documentation, getting the patches to Cakewalk, and waiting for their graphics people to create the box art. I see these patches as being useful not just due to their pure, straightforward sound, but also for being able to serve as the basis for additional programs in the future. If people like this expansion pack, it’s very likely I’ll make additional programs available at some point.

Meanwhile, I’m very happy with the way these patches turned out—and if you buy the Gibson EB-5 String Bass expansion pack, I hope you are too. They’ve already become my default “go to” bass sound, so in the end all that tedium was worth it.

[Note: The first example shows the EB 5-String Bass Expansion Pack Tapped Bridge instrument in a full musical context.]