Review – Acustica Audio’s Nebula
What if you could sample the character of FX units or signal chains the same way that you can sample instruments and recordings? Acustica Audio’s Nebula is a product designed to do exactly that.
by Per Lichtman, Nov. 2013
The Nebula Concept
What if you could sample the character of FX units or signal chains the same way that you can sample instruments and recordings? Acustica Audio’s Nebula is a product designed to do exactly that. Back when I first tried the product years ago, the potential for the platform was hinted by in the bundled library. But there are now around a dozen third-party Nebula library developers listed on Acustica Audio’s site with over 200 libraries available for sale at the time of writing, running the gamut in price from free to a few Euros to premium bundles over the 100 Euro mark. So now you can really hear what Acustica Audio Nebula can do.
The Short Version
This is going to be a long review because Nebula is very different from other products, so I’ll skip ahead to the good part for a moment: if you are using Nebula the way it is designed, then it can get you closer to the character of the sampled hardware than any other product I have tested so far (and there are a ton of sampled hardware libraries you can buy). That’s why it gets used on so many tracks I work on now. What’s it not designed to do is deal well with really loud input signals (you will need to turn down the output on a lot of soft synths) or work at the tracking stage (it adds a little latency). To get the most out of using it, you will want to use it either during mixing, mastering stages – or alternately using batch processing at any point after tracking, or during sound design. If you find yourself buying (and wanting to use) all the different 3rd-party libraries, you’ll also need a lot of memory and hard drive space, and optimally a fast processor. Several of the 3rd party libraries (including several from Cupwise and Gemini Audio) have been future proofed by including high CPU usage modes that currently require offline rendering, but that could be real-time capable on future systems. Now let’s get a bit more in-depth.
Nebula’s Strengths and Workflow
Out of all the things Nebula can do, perhaps the simplest to understand is the ability to “color” a signal. When dealing with outboard gear, every addition to the signal chain colors the sound somewhat, often in very subtle ways. When working entirely “in the box” on a computer with algorithmic manipulations, this is a nuance that sometimes gets lost. Nebula is not a traditional algorithmic plug-in. Algorithmic plug-ins have to create models to approximate the signal behavior and the complexity, quality and behavior of these models vary greatly, not only from developer to developer but also over the course of time for the same developer, relying on the developer to differentiate between important and unimportant variables. By contrast, Nebula provides tools for 3rd-party library developers (and all of its customers) to sample a signal path to record the exact characteristics of the way a signal gets modified in a given chain, within the limitations of the platform (more on that later) and to then be able to apply it to anything else. It really is an “FX sampler.”
By contrast to normal convolution techniques, Nebula sampling can keep track of the differences in the way a signal is processed at different input levels and it even supports modulated effects. It can smoothly interpolate between different settings in the hardware that is being sampled or different positions if physical space is being sampled via microphones instead. The technology is great at handling the sound and behavior of preamps, microphones, consoles and other gain stage and color devices, but over the years it has also gotten much better at handling dynamics as well – to the point where I sometimes choose it over expensive algorithmic compressors in my studio for certain tasks as well.
It takes a bit of adjustment to get used to it (EQs are normally sampled and operated a band a time; metering varies greatly from program to program; etc.) but it’s incredibly effective at both strong and subtle coloration and with an incredible array of colors.
Getting Started With Nebula
The key to figuring out whether Nebula is right for you is to set the thing up to get an accurate impression from the get-go. In writing this review I spent time talking with Acustica Audio, the 3rd-party library developers, experienced Nebula users and scouring the forums to see what sorts of experiences people had (both positive and negative). I compared their own experiences and suggestions with my own controlled A-B testing and came to a few conclusions about how to quickly get a sense for whether Nebula is right for you.
First of all, here’s a very common mistake that people make when using Nebula: they assume that it can be accurately evaluated by using it just like their other plug-ins. Nebula is in some ways more like outboard analog gear in that it’s very level dependent, but unlike that gear, it is simply not designed to handle overloading well. Let me repeat and re-emphasize that: you cannot judge the sound of Nebula by sending it a signal that overloads the plug-in. Nebula is designed to be used with headroom leftover (just like an awful lot of traditional hardware mixing), not to smash the signal up and past -0dBFS. If you do that, you go outside the range of the model and you’ll get horrible artifacts that don’t have anything to do with the sound that’s being modeled. There are a ton of plug-ins that are designed to handle sort of workflow (we’ve got upcoming reviews of URS Saturation, Studio Devil VTP and SoundToys Decapitator as a few examples) but Nebula is designed to excel at signals with a normal input level (and output level) range.
But if you aren’t supposed to max out the input and output levels in Nebula, how do you get a thick sound out of it? First, you need to pick a preset designed for the sound you like. Most of the 3rd-party libraries I was provided with came with some information about what the optimal input levels were, but generally speaking I found getting my input level peaks to land between -18dBFS and and -12dBFS provided consistently good results. You could get even better results for some libraries by optimizing the levels further and some libraries were designed to get an especially thick sound with peaks hitting in the -12dBFS to -6dBFS range. Long story short: when in doubt, make sure you’ve got at least 12dB of headroom and you are unlikely to run into problems but you can get more technical if you want (and VU style meters can really help).
Once you’ve got the signal being processed in the optimal range within Nebula, you can boost the output level with a fader (or fader plug-in) afterward if you need it to be hotter in the mix. So, here are two abbreviated cookbook approaches:
Workflow for Algorithmic Plug-in Designed for Hot Levels: (Optionally Attenuate Level to allow boosting more with plug-in)>Process with Algorithmic Plug-In
Workflow for Nebula: Attenuate Level > Check Meter> Process with Nebula > Bring Volume Level Up Outside Nebula
With that out of the way, let’s start with one of the best Nebula presets. The library included with the commercial Nebula versions should be thought of as the “General MIDI” of effects hardware sampling. In other words, there is a huge variety of sounds (and some of them are quite usable on their own) but almost all of them pale in comparison to the dedicated libraries you can get for the platform now. Much of the sampling for the library was also done at a much earlier stage in Nebula development so they don’t take advantage of the latest features. I mainly reach for the sounds in the bundled library when I can’t find them in any of the additional ones.
For example, in the Tape section of the bundled library, you’ll find presets labeled Tape 5042. These are sampled from a Rupert Neve 5042 unit, which you can also find in AlexB’s TSX library (along with other units). If you use A-B processing with the bundled library vs. with AlexB’s, the difference is quite striking: AlexB’s is obviously superior.
If you want to hear Nebula’s strength in reverb for free, I suggest going to http://www.vnxtsound.blogspot.com/ and downloading the free 2.5 second EMT 140 Plate program from the full library. Note: this sampled from an actual real plate, not a hardware box or plug-in. Here’s how to make it run correctly once you install it.
– Open the Nebula3 Reverb plug-in in your DAW.
– Go to the MAST page.
– Click next to “MODE” where it says “0 Simple” and drag upward until it says “1 Guru”.
– Click the second to last parameter on the right list that says “DSPBUFFER” drag it upward to 8192.
– Click the “Save” button inside Nebula in the upper right corner of the LCD display.
– Remove the Nebula plug-in from your session and reload the plug-in.
– Click on the word “Init” then navigate through Reverb> VNXT>EMT 140 Plate and click on the 2.5 second program.
There are more great plate sounds available but this is really is a good indication of just how high quality reverb can get on Nebula.
So why do I love Nebula so much? Because of the combination of variety and quality of the sonic possibilities it offers. It’s about being able to quickly compare the sound of gear that was sampled from across the world (some of it one of a kind) and having the models sound so excellent. It’s about having both really subtle and really strong coloration available equally easily and having so many developers releasing new libraries for the platform all the time – many of them quite inexpensive. In essence, I can buy a Nebula library for almost any occasion.
Nebula can be beautifully subtle but it can also be very dramatic when you want it to. Let’s take a look at a couple examples of getting strong coloration with Nebula.
Need to make a pristine digital virtual instrument synth sound more vintage? Throw a strongly colored synth preamp program on it, like one sampled from an Oberheim or Moog by SigntalToNoize. Or if you wanted a bit little less grain and vibe but a bit more exaggerated bass you could use the pre from the Korg MS20 that AlexB sampled in Vintage Synth Filters. After that you have several options for adding additional coloration. Do you want to capture the sound of analog tape or a cassette using an appropriately named program from SignalToNoize, CDSoundMaster or CupWise? Or would you rather capture the sound of one of the effects units in AlexB’s TSX (which are designed to warm up the signal in various other ways)? Or would you prefer to explore the sound of the preamps or transmission approaches of using a radio using one of CupWise’s innovative libraries? There are tons of compelling possibilities at your fingertips and by setting up a couple tracks with a Nebula instance on each, you can quickly compare and contrast the results with each in a way that normally takes much longer to do with actual hardware.
Here’s a quick primer on the amount of coloration with aforementioned tape libraries or the tape style FX in AlexB’s TSX (which you should probably skip if you’ve used actual tape). When programs mention IPS (inches-per-second) it has to do with the speed at which the tape goes by during recording and playback. The over-simplify greatly, the higher the speed is, the more “accurately” the high frequency content gets recorded. Lower speeds will sound more obviously colored than high speeds, so my suggestion is to start by auditioning the lowest speed offered for the unit you have in mind in order to really be able to hear the coloration, before trying to find the optimal speed for your content. All the units that supported 7.5 IPS or slower were able to color the sound in an obvious way if desired – 15 or 30 IPS tends to be good when you want less obvious coloration.
For more subtle coloration, you can load up a high quality modern console or preamp. Such gear was designed to maintain signal integrity using modern techniques so the coloration is often much more subtle than with vintage gear – but no less important.
As one top coach said: “Shaving a few tenths of a second off the 100 meter dash for a beginner won’t help much. But if I can do that for an Olympic athlete, it can make the difference between a good effort and a medal.” When it comes to using subtle coloration, it won’t make a bad mix sound good. But it can be used to help take a sound or a mix and push it right over the edge from “really good” to “nailed it”. It can also make the process of mixing easier.
Included vs. Additional Libraries
The majority of users are more likely to buy Nebula for the libraries currently available than to try to roll their own. That is why the huge collection of libraries currently available for sale (as well as several additional free ones) is so critically important. Across the board, I found developers had taken advantage of their greater experience with the sampling process (and the evolution of the Nebula engine) to create progressively more powerful libraries over time. In other words, it’s often best to start by looking at a developer’s most recent libraries to get a sense for what they are capable of.
Some of the easiest libraries to use are the preamps, consoles and microphone libraries. With these libraries you can essentially load the program and get “the sound” without having to tweak settings. The large catalogs assembled by developers like CDSoundMaster, AlexB, Henry Olonga and SignalToNoize each include many types of libraries, while many other developers currently cater to different niches. For example, ranscendingMusic’s libraries (http://www.tmusicaudio.com/nebula.html) cover various types of specialized outboard gear, from enhancers and stereo field modules to a saturation library made using the Thermionic Culture Vulture (one of the same units modeled in algorithmic fashion in the competing SoundToys’ Decapitator, which we will cover in an upcoming review).
The bundled microphone library is outdated and unrepresentative: the 5 kernel versions here cannot compete with the 5 kernel versions in top libraries, let alone the “full” ones that average 10 kernels. I would suggest not even downloading it.
The three microphone libraries currently available from http://www.gemini-audio.net/ have great sound quality, are easy to use and don’t take up that much space (the zipped download is less than 60MB per model). While I’ve talked a lot about subtlety there’s nothing subtle about the difference between these models and the ones in the library.
Henry Olonga’s http://www.nebulapresets.com/ also features a great collection of high quality microphones as well.
Between Henry Olonga and Gemini Audio, the Royer microphones are especially well represented and have the highly desirable ribbon color.
Perhaps one of the most appealing applications for Nebula is the collection of console libraries. At time of writing, four third party developers have created console libraries: Alessandro Boschi (AlexB), Analog in the Box (AITB), CDSoundMaster (CDSM) and SignalToNoize (STN). Surprisingly, there appears to be no overlap in the models between them, meaning users currently can buy over 15 consoles (18 if you count the EQ/console combo options) to have at their fingertips. There are only a handful of companies outside Nebula sampling that have tackled the “virtual console”, and even they would have to acknowledge that the variety available on the Nebula platform is unmatched by any other hardware or software plug-in to date.
Here’s a bit of info on the consoles, taken from the unofficial (and unaffiliated) list at http://www.learndigitalaudio.com/blog/nebula-vst-universal-program-explorer – the accuracy has not been verified and the names are generally not advertised by the developers. AITB’s Tube Console Bundle emulates an unspecified custom console and STN’s offering comes from the sound of the 1960s American Langevin AM4A (or L-401). AITB’s Tube Console Bundle emulates an unspecified custom console and STN’s offering stems from the sound of the Langevin AM4A. CDSoundMaster offers 5 consoles (including ones from Trident, Sphere, BBC, MCI and an unspecified British manufacturer) as well as two EQ (an Amek 9098i and another which a senior forum member indicates stems from Orphan Audio Hardware).
The largest number of consoles comes from AlexB: two Neve consoles, another Rupert Neve Designs one, two SSL desks, an API, TL Audio and even the Neumann console model used to master Pink Floyd’s Dark Side of the Moon.
As an audio engineer, it really is difficult for me to say enough good things about these specific libraries. High profile studios generally don’t have more than a few of these (some rely on one) and here are over a dozen that you can run on your computer without additional hardware, in one place, without having to travel, and hear them in seconds. I was a little skeptical as to how big a difference these would make – I thought it might be subtle stuff that only audiophiles could hear. Well, maybe you won’t hear the difference right away if you are listening on laptop speakers in a noisy room – but play these on any decent set of speakers (or even just a $200 pair of headphones plugged into the lowly headphone output on a few year old MacBook Pro) and the difference made by adding just the mixbus/direct out/master output stage to mix before mastering is very noticeable. The difference only gets more dramatic as you start to apply it as intended.
Each console include programs to use per channel (generally suggested as the first FX but useful anywhere in the chain) on each instrument and another to be used on the master bus. Some of them, such as AlexB’s Vintage Blue Console (VBC) and Modern White Console (MWC), go even further. CBC and MWC provide 6 or more different line input channel programs, two microphone input programs, multiple variations of the final output stage coloring (often through tube switching), a separate sampling for the FX send-return bus the console used and several additional programs intended for user on bus/group for a given collection of instruments (such as the synths and pads or the drums, etc.) The cumulative effect of applying these stages to the different parts of the mix can be really dramatic and I was honestly surprised by the sheer depth of the sampling in many cases. Here’s a suggestion for how to get a sense for it: apply just those effects as intended to every part of the mix, adding no other FX and then render the mix both with and without the FX. Normalize both and listen. In my experience the biggest difference came in the sense of space and the tonal character, more so than in terms of volume or “punch”. Try doing the same thing with two consoles of vastly different characters, like CDSM Globe console and AlexB German Mastering Console. I knew there would be a difference, but I didn’t expect for it to be as great as this.
I find that when using the console programs on raw recordings, I don’t need to do as much EQ work and that the tonal balance often seemed to be easier for compressors to handle. I did many A-B tests experimenting with different FX orders in my signal chain just to be sure, and sure enough, the console programs most often sounded better before the compressors (or other dynamics processing) were engaged.
Each of the libraries sounds different and has a character of its own, with my colleagues demonstrating repeatable preferences in blind testing. From the “radio-friendly” sound of AlexB Classic Logic Console and Modern Logic Console, to the far more relaxed highs and sense of space of CDSM’s Trident to round (and noticeably less colored) sound of AITB’s Tube Console Bundle.
I’d like to take a short moment to dwell on STN’s offering. Eric Beam at STN’s L401 sampled really sounds quite different from the other consoles. I found it to be very useful on string quartets, for instance, that had too much air or rosin noise or had somewhat exaggerated dynamics. On drums, it makes things less bright and splashy. It can be great for taking some of the “digital edge” off drum machines, for instance. It’s useful for many other things too. Essentially, it feels like the soundstage is both a little closer and a little smaller, with bit less dynamics and a smoother sound. It’s my “let’s simplify, reign things in and focus a bit” console choice. Warm, smooth and focused
Delays (much like reverbs) require a larger DSPBUFFER setting than most other program. In my testing I used 8192 samples, but I suggest experimenting with different values to find the one best suited to your system. Note that running your session at a higher sample rate means a proportional increase in CPU.
In addition to Cupwise’s feature-rich “Pioneer Analog Echo / Reverb” library (discussed in the reverb section), there are currently three dedicated delay collections for Nebula, all offered by STN at prices ranging from $18 to “free”. Once installed, they are found in the Delay (DLY) menu inside Nebula. Each is full of character and brimming with vintage flavor – as well as vintage eccentricities. They are light on controls (dry and one or two FX level controls mainly) but you’d be hard-pressed to find recreations of a Maestro Echoplex or ADA TFX4 in DSP anywhere else. The MoogerFooger Delay has been emulated elsewhere before, but rarely with such a “warts and all” approach to the detail of the sound. Throwing one of these into a mix is like opening up a time-machine to a different era. Levels can be tricky (after all the original gear used included feedback) but it’s worth spending the time to get it just right. My advice – when in doubt, cut the input level and boost the output level (even if you have to do so outside the plug-in to avoid clipping). The Plex (Tape Delay) is taken from a Maestro Echoplex with programs sampled at various delay times, with feedback. The programs sounded great to me, full of a vibe and character that’s quite different from normal tape-delay plug-ins. If you’re mainly looking for tape delay with lots of vibe as opposed to an exact timing beyond those covered in the presets, this is the one for you.
The downside to Plex (as compared to a more modern plug-in that might have less character) is that you must give up niceties like tempo-sync, feedback control, or even basic timing control. Instead you have 7 presets. Honestly, I have yet to hear a more convincing vintage tape-delay anywhere else and if it had those niceties, it could probably supplant all my existing tape delay plug-ins. But luckily, the presets cover a range of timings: 439 ms, 380 ms, 312 ms, 245 ms, 180 ms, 123 ms and3 ms.
The Plex programs could act up at times on my review system with unintended feedback building up randomly, but hopefully it’s an isolated (or easy to fix) problem.
The “Analog Doubler” library comes from an ADA TFX4 machine, is offered in three settings and does pretty much exactly what the name would lead you to expect. I didn’t really think I needed a doubler but found this one quite effective at thickening up voice and drums. It’s a lot better than I expected from a free library.
I didn’t receive a copy of the MoFo delay in time for this review so I will try to cover it in a future issue. From the demos I heard, it seemed to continue STN’s attention to capture the vibe of the original gear.
At first I was slightly disappointed with the filters selection for Nebula. After all, the offerings in the bundled library didn’t compare favorably to plug-in alternatives and none of the commercial libraries I tried were capable of the screaming resonance of my favorite plug-ins like Cytomic The Drop. But then I noticed what I’d been missing: filter preamps. STN and AlexB both offer a collection of filter preamps that work well with any other filter your throw at them – and by their very nature, they often have a stronger color than a microphone preamps. STN offers several different ones (Moog, Oberheim, Arp) as well as preamps from various other synth-related gear (drum machines, etc.) that are all quite inexpensive. AlexB’s filters are included in the Vintage Synth Filters collection.
You can also find an exotic Urei filter at http://www.cupwise.com/cup/ that is incredibly smooth and has a very gentle slope (as well as being able to get some really crazy modulation effects, etc.)
More on the Libraries Soon
From AlexB’s variety of well-known consoles, to CDSoundMaster’s range mixture of more unusual and one-of-a-kind ones (as well as the Trident sound associated with some of Michael Jackson’s top records) to the eclectic collection of gear in the SignalToNoize archive (especially interesting to anyone looking at vintage synth and FX gear). From the “pushing the envelope of what a Nebula program can do” approach in Cupwise’s releases (which often add parameter controls not seen in any other Nebula programs) to the amazingly comprehensively sampled API 2500 compressor library offered by Gemini Audio. From Gene Lennon’s faithful recreation of some of his most used TC System 6000 presets (which sound eerily close to the original from the comparisons he provided), to the outboard reverb emulations by STN, to the multi-position and mutli-mic Theater of Life by Roomhunters, to the plates from VNXT, Cupwise and Tim Petherick. From Ownhammer guitar programs (which I used to great effect on the sessions by the excellent Jimmy Haun) to TranscendingMusic’s range of processors and FX units (including great stuff for free). From the excellent Silk EQ by Tim Petherick, unique Manley sound in AITB’s Mammoth EQ to the air EQ presets from Henry Olonga. Not to mention the array of microphone models offered by Henry Olonga and Gemini Audio… there’s an overwhelmingly huge selection.
We’re coming back for more in the coming months, but here’s a list of the developer websites and at least one of my favorite libraries from them.
AlexB: German Mastering Console (GMC) and Vinylizer (VNL). Keep your eyes peeled for the upcoming Modern Flagship Console (MFC).
http://www.analoginthebox.com (should be back soon)
AnalogInTheBox: Mammoth EQ
CDSoundMaster: Apex Tape, R2R, TapeBooster+
CupWise: Custom Plate, Smooth 609A BusComp (off a Neve 33609A), Urei emulation
GeminiAudio: API 2500 Compressor or Royer 122V microphone library.
Gene Lennon: TC System 6000 reverb collection (GLP)
Henry Olonga: Microphones (almost any but definitely the Royers) and the EMI preamp. Keep your eyes peeled for their free upcoming Bricasti M7 reverb library in conjunction with CDSoundmaster.
Note: Part of the proceeds from the libraries sold at NebulaPresets.com go to supporting the Mumvuri Project, an orphanage in Zimbabwe.
OwnHammer: Any of the Studio Mix Libraries.
http://roomhunters.net/ (Currently unavailable but Acustica Audio is working to host this in their shop soon)
RoomHunters: Theater of Life multi-position real life reverb collection.
SignalToNoize: Oberheim and Moog Preamps, Plex Delay, Analog Delay
TimPetherick: Silk EQ, ELC24 V2 EQ, Spring (SSP)
TranscendingMusic: VCult (from the Thermonic Culture Vulture)
VNXTSound: EMT 140, BX20
Zabukowski: Creator of Nebula Setups 2 and NebulaMan utilities for Nebula – not libraries.
[Update 2014: Acustica Audio has added another official 3rd party developer, Lars Rüetschi, whose guitar and bass oriented Nebula programs can be seen at http://www.thesessionguitarist.com/nebula-programs.html but I have not had a chance to audition them yet.]
Come Back Soon
It should be noted that during the course of this review, two of my favorite Nebula developers were in the process of making changes that mean their libraries are temporarily unavailable for sale. [Update: Acustica Audio has stated that they are in the process of trying to make 3rd party libraries, including these, available for sale through the store at http://www.acustica-audio.com/, so check there as well to see if they become available.]
First, AnalogInTheBox.com – one of the oldest 3rd-Party Nebula developers, founded by Niklas Richter (formerly known as Velinas, the moderator of a very popular forum that used to create and provide cool new Nebula programs before the creation of Analog in the Box). While Niklas still provides access to the site for existing customers and is in the process of finding someone else to take care of the site and brand so that people can start buying the products again. I hope this happens soon as the challenge and response system for the libraries presents certain issues at present (like limited authorizations and some compatibility issues with newer Nebula versions) – something that wouldn’t matter if libraries like Mammoth EQ weren’t so good that I wanted to be guaranteed to be able to use them. In fact, Mammoth EQ is both uncommon and one of the best sounding approaches to the Manley Massive Passive currently available anywhere (with only 3 other emulations existing on any platform that I’m aware of). So stay tuned for its return.
At the time of writing RoomHunters.net had their site temporarily unavailable, but they plan to return soon – so my suggestion is to pretty much just check them every day until they came back. 😉 What RoomHunters.net offers with Theater of Life is unlike any other Nebula library and has few peers in the reverb world in general, sharing the most in common with something like VSL’s much more extensive and expensive MIR. To be clear, Theater of Life does not sound like a large concert hall: it sounds like a small theater and the miking is more on the “full of character” than the “pristine and clear” side of the equation. It has very low CPU usage compared to other Nebula reverb libraries (and actually relatively low for reverb in general with the right settings) and uses around 10 microphone programs (one of them mono) with panning and distance controls that take full advantage of Nebula’s interpolation capabilities to give you real-time automation capabilities for distance and placement. I definitely encountered some phase issues, but the library is extremely inexpensive and on the right material (or for something that just needs an authentic small theater vibe) the result is compelling, realistic, low-CPU and easy to use.
The Different Nebula Versions
There are many Nebula products now available, from the core product range (Nebula 3.6 Free, Nebula 3.6, Nebula 3.6 Pro, Nebula 3.6 Pro Server Bundle) to standalone products using the so-called “Acquavox” Nebula engine (like the recently released “Stradipad” collection). So the question is: which product does what? Which one should a new user be looking at? We’ll work our way down from the top to make it simpler. As a quick note, the products are all labeled in the Nebula shop with their full version number only, but since the 3.6 release is quite different from the 3.0 release I originally encountered years ago, I will refer specifically by version number here. Note that this review was written using the Nebula 3.6 Pro Server Bundle.
At the top, there’s Nebula 3.6 Pro Server Bundle (currently €189.00). This is the flagship model and will run every Nebula library released with the highest performance and highest memory efficiency currently available, with support for Windows and Intel Mac OS X in 32-bit and 64-bit. Unlike the other paid versions, Nebula 3.6 Pro Server Bundle consists of two parts: the normal plug-in and a server application. Since the bundle comes with multiple licenses, the server application can be run from a different computer than the plug-in (as long as you start it before loading the plug-in), allowing you to use additional resources on another computer. I did not have occasion to test this during the review process, but did ensure that the server application functioned properly when run locally. When the plug-in is used without the server application, it simply functions like a souped-up version of Nebula 3.6 Pro. Both Nebula 3.6 Pro Server and Nebula 3.6 come with the N.A.T. 3 sampler application to sample your own hardware for use with Nebula. The next step down is the Nebula 3.6 Pro Bundle (currently €139.00). During my testing, all libraries I encountered supported Nebula 3.6 Pro (though some did not support the less expensive Nebula 3.6 Bundle). In other words, this is the least expensive version that will give you access to the full Nebula range on both Windows and Intel Mac OS X in both 32-bit and 64-bit. Since Nebula 3.6 Pro comes with two licenses, the main difference compared to the more expensive Nebula 3.6 Pro Server Bundle is that you don’t have a server/local server application you can run on another computer and there is no memory sharing code. This could be thought of as the version with the most bang for the buck and is the lower price-point at which you are assured the “full” Nebula experience. As mentioned earlier, it comes with same effects sampling hardware as 3.6 Server (N.A.T. 3) .
The least expensive member of the core product line is the Nebula 3.6 Bundle (currently €79.00). This version supports most 3rd-party Nebula libraries, but read carefully before you buy one because some require either Nebula 3.6 Pro or Nebula 3.6 Pro Server (though such libraries are currently in the extreme minority). This version does not have the broad OS or 64-bit support that the higher-priced versions offer (it’s Windows only, for starters) and comes with 1 license instead of 2. It also comes with the less powerful N.A.T. 2 compared to N.A.T. 3 offered in other versions. The higher-priced versions also use a superior Nebula engine. While it was slightly difficult to sort out exactly which aspects sounded/performed/were supported better in Nebula 3.6 Pro vs Nebula 3.6, Acustica Audio’s comparison table mentioned both Dynamic Range and Multi Dimension Models. What makes up the “paid engine” products currently available in the Nebula range, leaving Nebula 3.6 Free and the standalone libraries off the AcquaVox engine. For this review, I spent almost no time with AcquaVox and the standalone library products that don’t require users to own a Nebula product. However, I can say that Nebula 3.6 Free appears to be largely based off the code base of the Nebula 3.6 Pro Server, but cannot load most libraries (not even the library bundled with – only ones specifically designed for Nebula Free – not unlike Native Instrument’s approach with Kontakt Player as opposed to the full Kontakt version). It comes with a much smaller library and a few developers provide patches that can be loaded into it, but Nebula Free might best be thought of a way to get a sense for the platform before spending money on it.
As a legacy note, while writing this article, Acustica Audio discontinued the Nebula 2 series (an inexpensive but both more difficult to use and less powerful version). This was a good move for the brand in my opinion since having too many products could get confusing for new users. On top of that, several libraries already sounded noticeably better in Nebula 3 (let alone 3.6 Pro or Pro Server) if they were even supported in 2 at all. In other words, the odds of people getting an accurate first impression are higher now.
Zabukowski Software’s Nebula Setups 2 and Understanding Nebula Files
Nebula libraries primarily consist of two types of content: Programs and Vectors. Programs are smaller files that contain parameters and metadata (much like “programs”, “presets” or “patches” on traditional samplers). Vectors are larger files that contain the actual sampling data that Nebula references during processing. Both types have their own named folder in the Nebula folder. Here are the default locations for the Nebula system folder on OS X and Windows 7 systems.
Windows (where “C:\” is the system drive):
If you’re using the default Nebula settings, you’ll notice that the more Programs you get, the longer Nebula takes to load. In addition, you may start to find that it takes longer to select or find the Programs or to organize them effectively. Different programs may also require different Nebula configuration settings to make the most of them (for instance reverb Programs often require a higher DSPBUFFER setting inside Nebula than preamp Programs).
Zabukowski Software (http://zabukowski.com/software/ or available for purchase at at http://www.acustica-audio.com/index.php?page=shop.browse&category_id=67&option=com_virtuemart&Itemid=53) has created Nebula Setups 2 to address all of these issues.
Nebula Setups 2 uses a graphic interface to create more optimized Nebula configurations. If you read the PDF called “Nebula Setups 2 Quick Guide” and go through the “Install” and “First start and basic setups” sections, you can start creating custom setups of your own.
Here’s how I created one optimized for Programs with a long tail (like some of the reverbs and delays).
1) I started Nebula Setups 2, went to the Tweaks menu and selected “Add Parameter”.
2) I scrolled down to the “DSPBUFFER – Internal Buffer Size” parameter.
3) In the value field I typed in “4096” to increase the buffer size to 4096 samples and clicked OK.
4) In the upper left window I selected the “REV (Reverb)” category.
5) I right-clicked on the category and selected “Add program(s) from selected category”.
6) I repeated steps 5 and 6 for any other category with long tails (in my case the “DLY (Delay)” and “SSP”).
[These categories came from programs from SignalToNoize.com and TimPetherick.co.uk but your own categories may vary.]
7) I went to the Setup menu and selected Save.
8) I typed “Long Tail – Nebula” in the Setup Name field.
9) I unchecked the “32 bit (x86) VST plugin” box since I was only using the 64-bit version.
10) I clicked “Save”.
The next time I loaded my DAW, “Long Tail – Nebula” appeared in the list of plug-ins. The reverb programs (like the free VNXT EMT 140 2.5 second one) ran much better than with the default buffer used in my standard Nebula plug-in. The plug-in loaded much quicker too, since the index included just 399 programs instead of well over 7,000 like the main Nebula instance on my system (which also made it quicker to find the reverb programs I wanted to use).
The main key to the way Nebula Setups 2 works is the XML file approach that Nebula uses.
One XML file is located in the same folder as the Nebula plug-in and share the name of the plug-in (for instance “Long Tail -Nebula.xml” in my example). This XML file stores several Nebula configuration parameters (like the DSPBUFFER that sets latency) and tells Nebula what XML file to look for in the “Setups” folder.
The “Setups” folder is located in the Nebula system folder XML file is paired with another in the “Setups” folder within the Nebula system folder mentioned earlier. This folder also contains an XML file with the same name (once again “Long Tail -Nebula.xml” in my case) but this XML file is an index of the Programs that Nebula will display when you load the plug-in.
At the time of writing, a release candidate of Nebula Setups 2 was available and functioned properly in my testing under both Windows 7 x64 and OS X Mountain Lion 10.8.4. [Update 2013-11-21: It is now available for sale at http://www.acustica-audio.com/index.php?page=shop.browse&category_id=67&option=com_virtuemart&Itemid=53]
Zabukowski Software’s NebulaMan
NebulaMan is another tool you can find at http://zabukowski.com/software/, or for sale in the Tools section at http://www.acustica-audio.com/index.php?page=shop.browse&category_id=67&option=com_virtuemart&Itemid=53, this one designed to audition in real-time or batch process Nebula settings onto various audio files. It allows you to create several different Nebula combinations (stack as many instances in serial as you want), provides the option to automatically bring input levels to Nebula’s optimal range and even provides the ability to link input and output level sliders if you want. All of this allows you to very quickly set up, compare and process large numbers of Nebula files without opening a DAW, along with a few additional tricks of course. The process is made quicker by being able to use shortcut keys to switch the active Nebula chain being used for preview playback at any time.
As a point of clarification, the batch processing is based around processing several files in a row with the same settings, not processing the same files with a lot of different Nebula programs in a row (though processing files with any of the handful of setups you have open at a given time is a quick process).
Depending on the batch capabilities already offered in your DAW (Reaper being fairly extensive in this area for example) you may not need the batch capabilities on offer here. But the fact of the matter is that NebulaMan makes the process very quick and very easy and the small tweaks (like settings levels into Nebula’s optimal range at the push of a button with the option of whether to compensate output or not) is a very nice addition indeed that may make the software useful to you nonetheless.
Organization and Optimizations
I love the sounds I can create with Nebula and I love the array of possibilities and that the pricing of the libraries and software is very fair, but there are still things that could use improvement. Note that I’ve included certain issues with 3rd-party libraries here, even though Acustica Audio obviously is not responsible for some of the choices that 3rd-party developers make.
First of all, organization. The current way the programs are tagged and organized is inconsistent and confusing. For example, there where I could not find a library I had just installed, either using the Nebula interface itself or by using keyword sorting within Nebula Setups 2. Here are some of the reasons for that.
1) Sometimes 3rd-party developers create entirely new categories for each library; sometimes they put their programs into sub-categories of existing categories (like “Pres”); sometimes they use a single folder for almost all their programs (like “HO” for Henry Olonga) or they may create a new category for some of their collections (like “CCC” for the “CDSM Classic Console” libraries). Often the category is not specified either on the web page for the library or in the accompanying documentation making it more and more difficult to find new libraries in the list as your collection expands.
Potential fix for 3rd-Party Developers: please always specify the category your library will install with, preferably both on the web page and in the documentation. As a user, I also wouldn’t mind a more standardized approach to category organization, but that may be asking for too much. 😉
2) If we look at the fields that Nebula Setups 2 reads, we’ll see categories, “Name”, “Description”, “File Name” and “Number.” You’ll notice that there is no dedicated field for “Author”, “Library” or “Website” so the description may include one of these – or it may include an actual description. If you installed several libraries at once (sometimes I would install 10 or more in one day), you may have to go back to look at the file names used in the install just to be sure what library the program you’re loading came from (let alone where that library came from). This is not just an academic point: I came across several forum posts online where users were trying to find additional libraries by the author of an old Nebula library… only to realize that they had no way of identifying the author of the free programs in question.
Potential fix for Acustica Audio and 3rd-Party Developers: Add more dedicated tagging/metadata fields (“Author”, “Website” and “Library” should be the bare minimum) and leave the description field for actual description. If a user likes a given Program, they usually will want to know where it came from, and it makes it much easier for them to buy more if they do. 🙂
Second, there’s the issue of configuring Nebula for different types of programs. Each Nebula plug-in format installs two versions by default: Nebula3 and Nebula3 Reverb. But most developers suggest using Nebula3 Reverb for all their programs, regardless of content type, for best sound quality. So far everything is fine and it seems like Nebula3 is primarily there for legacy support. But then we start to get into files that require or benefit from different optimizations: delays and reverbs with long tails (which require a higher DSPBUFFER setting than normal programs) or a handful of specialized compressors (like some of the best libraries from Gemini Audio). If you get Nebula Setups 2 then it is great tool for improving load times and keeping Nebula settings optimized for different types of programs, but the fact remains that the process is still not as easy as it could be.
From an ease of use perspective, it would be simplest if 3rd Party developers could specify optimizations within the Program file so that users didn’t have to do any optimizations themselves at all. But as multiple people pointed out to me while I was writing my review, some parameters (such as DSPBUFFER) will have different optimized values on different systems. I generally found myself using identical values on two wildly differently spec-ed OS X and Windows 7 systems since I was most concerned with minimizing CPU usage – running almost exactly opposite to the workflow suggested by SoundOnSound in an earlier review. In other words, I can see Acustica Audio’s difficulty here. Maybe they can come with an elegant solution that I can’t think of.
In the meantime, perhaps one workaround would be to at least include a 3rd Nebula instance (perhaps called “Nebula3 High Latency”) that would be configured with a 4096 or 8192 sample buffer, so that new users would be able to run long-tail reverb and delay libraries without changing configuration settings. Because the first time I tried to load some of those programs using Nebula3 Reverb with the default buffer, they almost completely failed to playback, sending me searching for the tweak that would make them run (which I eventually found to be the DSPBUFFER).
In fairness to Acustica Audio, a significant portion of the existing Nebula market seems to be power users that either like or don’t mind making these kinds of tweaks. The sound of Nebula is so good, however, that a lot larger market might enjoy using it if the experience catered more to new users.
Nebula and 3rd Party Developer Websites
The modern Acustica Audio website is a huge improvement from what I encountered when I bought Nebula 2 years ago. It’s friendlier, faster (download speeds were generally excellent), more visually appealing and easier to navigate. Nonetheless, there’s still room for improvement – something that could also be said of many of the 3rd-Party Developer sites as well. For the Acustica Audio website, the biggest improvements would be:
– [Updated December 1st, 2013] Originally I had written down “Further automating/improving the challenge and response process so that users can get their Nebula license within seconds or minutes instead of hours or days.” Of course Acustica Audio decided to improve that, so the automated system should now provide users with their code in “3 to 60 minutes.” In my testing, it took less than a minute. I’m glad to see the company get proactive about this.
– Providing clearer descriptions of different Nebula versions on the shop page itself, rather than relying on users visiting the comparison table to find out.
– Keeping the F.A.Q. more frequently updated with the information that the Acustica Audio team provides elsewhere on the site in their forum posts. [The current F.A.Q. is up to date at the time of writing, so this has hopefully already become a thing of the past].
– A clearly highlighted “Getting Started” guide for Nebula, possibly with videos.
– Having the splash page include rolling news updates in addition to the currently featured graphics of new products.
For the 3rd-party developers, here are a few suggestions that would help to better cater to a different breed of potential customers. [Updated November 21st, 2013] Acustica Audio has said they are moving to start hosting 3rd the third party libraries in a centralized shop. This could address many of the issues mentioned here if it comes to fruition.
– Make it easier to buy lots of things at once. I am not talking about doing deep discounts – heck, you don’t even have to discount at all if you don’t want to since the prices are already very fair. I mean make it possible to add all the libraries you sell at once to the shopping cart with just a click or two. In one case I had to spend what seemed like over 30 minutes adding libraries one by one to the shopping cart to buy everything from a given developer. So – if your content is good and someone just wants to buy all of it at once, make it easy for them to do that. If you want to do a discount too, that’s fine – but mainly just make it easy to buy a lot of things at once.
– Organizing your shop. With very few exceptions, the shopping categories were not as clearly organized as they could be. Issues included only allowing to sort by overly specific categories (as a user I would rather be able to see several libraries on one page than have categories so specific they include just one library) or almost entirely omitting categories at all. Since some of the developers use a blog style interface, it can also be difficult to get an overview of what programs are available. Any improvements in this area will be greatly appreciated by your new customers.
– More tutorials, video and audio examples.
A Note About CUDA
I’ll briefly mention CUDA support. Nebula’s CUDA support has been mentioned in the past (and still gets listed on Acustica Audio site in the Feature Comparison table for both 3.6 Pro and 3.6 Pro Server.) At the time of writing, CUDA support was not documented with sufficient consistency for me to recommend exploring it and many posts on the topic were discouraging. Contrasting answers were given on the topic on different parts of Acustica Audio’s site and there’s a general sense that this has not been developed to the point of being a marketable feature. This did not adversely affect my opinion of the product but I wanted to clarify the point for any readers considering the platform on the basis of CUDA support.
Providing any sort of a comprehensive look at the Nebula platform extends beyond the scope of a single review (even one as long as this one) so we’ll be going much further in depth with specific Nebula libraries and standalone products in our next review. But the most succinct way I can put it is that Nebula platform is one of the most exciting mixing, mastering, sound design and processing tools that I’ve ever encountered in digital audio. When it’s used to sample analog hardware it usually gets closer to the original sound than any emulation/sampling I’ve heard – and the array of libraries available for it is extremely compelling. Factor in that you could buy every library available for the platform today (even a la carte and without having to use a single discount or bundle) for less than the price of Waves Mercury or the UAD Ultimate bundle, and you have a very compelling challenger in this area. One of the artists I mix for (that has experience with both the aforementioned products) even says that hearing the sound of Nebula on being applied to her mixes is her favorite part of the whole mixing process.
It’s not for everyone of course (especially not people that need low-latency real-time effects while tracking or with slow CPUs, low ram and small hard drives) and there’s still room for improvement in terms of the user interface (making it easier to display and modify multiple bands of EQ; clearer and more consistent metering for compressors, etc.) but it really is one of the very best digital options available anywhere for adding vibe to your sound. And it’s getting better every day.