How To Download Free Products
Purchasing and upgrading questions
1 I have purchased Nebula. How do I change my user account to 'Customer'?
It takes a little while for customer accounts to be upgraded. Please be patient and bear with us! It may take up to a day for registration. You can check if your account at Acustica Audio website has been upgraded by looking at the color of your account name on the forum. If it is pink, then you are a customer and should have full access to the download section.
Yes. You will only pay the difference for the upgrade price. Keep in mind commercial Acustica Audio products share the same emulation preset library.
The are 3 'levels' of Acustica Audio products. Commercial products (Nebula2 and Nebula3/Nebula3 Pro) are available from the official Acustica Audio website. Limited editions are specialized versions with a limited amount of emulation presets. The limited editions (LE) are as follows:
- Nebula3 CM, available with select volume(s) of Computer Music magazine (more information at http://www.computermagazine.co.uk/).
The free version is a demo version. It is similar to the limited editions, but you can download it from the official Acustica Audio website for free. Both the limited and demo versions have a smaller amount of emulation presets. While you can acquire all the different varieties of Nebula and they will all run together on one PC, Nebula3 Pro contains everything - the plugin, GUI, full comercial library and latest version of the NAT (so you can create your own Nebula presets!).
What are the different versions of Nebula LE?
Acustica Audio has created some special editions of Nebula for various computer music magazines and websites.
Nebula3 KM has 10 exclusive programs. (exclusive for the keys magazine, you'll find these in the Nebula commercial library too)
Nebula3 DR has 20 exclusive programs, different from CM and KM. (exclusive for Direct Resolution custom DAWs, you'll find these in the Nebula commercial library too)
What are the differences between Nebula bundles?
Nebula3 Pro Bundle
Free bug-fixing and support.
Free bug-fixing and support.
Nebula3 FREE Bundle
Nebula License/Key questions
1 I have bought the plugin. What should I do next?
You will need your account registered on the forum so you can be upgraded to a Customer. It takes a little while for customer accounts to be upgraded. Please be patient and bear with us! It may take up to a day for registration. You can check if your account at Acustica Audio website has been upgraded by looking at the color of your account name on the forum. If it is pink, then you are a customer and should have full access to the download section. You should also send in your Nebula3.ser file (located in your Nebulatemprepository folder - specified on the original install of Nebula) to email@example.com as part of the registration process. You will need to run Nebula once after installation for this file to be generated. We will then send you a registration file that, once run, will register Nebula as the version you have purchased.
Please post on our forum and we will sort things out for you ASAP!
Nebula is not registered correctly. Please check that you have sent the correct .SER file to firstname.lastname@example.org. The .SER file should be from the computer you intend to use Nebula on. Don't forget to make sure you are installing the key in the Nebulatemprepository folder (specified on the initial Nebula install). This is a common problem, so check that the right folders and files are being used. If all else fails, don't panic! Post on our forum and we will do our best to help you.
You are using the wrong key. Check through the information in the above question #3 to ensure you have sent us the right information!
Changing the motherboard or the processor will not change the Nebula key, so don't worry! The only time you will need to request a new license is when you change your system hard drive (the one that Windows is installed on). If you have changed your system hard drive please send us an e-mail at email@example.com.
Nebula3 Pro will enable you to use 2 licenses. Nebula3 and Nebula2 only contain one license each. So if you want two copies of Nebula, please ensure you purchase Nebula3 Pro.
Download and decompressing files questions
1 I have registered, but when I click on the download link I get the following message: "File access denied".
Your are not logged into Acustica Audio forum. Go to the Acustica Audio main web page at http://www.acusticaudio.net/, click on the forum and click on log in. Enter your nickname and password.
If you have the security settings on your web browser set too high, it is possible that the cookies required to maintain a session with our forum and download system are not being stored or created correctly. Please consider temporarily changing your security settings in your browser. To the best of our knowledge there is no adware or spyware or any other malicious code at work on our servers and Acustica Audio does not use dataminers or other such technology to infect or afflict our staff and customers.
2 Where can I find NebulaSampler on your site? Is it free?
NebulaSampler is a part of the N.A.T. Tools - it is free for Nebula3 customers. Please note that N.A.T.2 is available with Nebula3, and N.A.T.3 is available with Nebula3 Pro.
3 I downloaded the main installer but what are all those other libraries for?
Nebula is a multiFX processor - the installer setup only installs the VST plugin in order to use emulation presets. The emulation preset libraries contain all the presets that you can use with Nebula (compressors, preamps, eqs, reverbs, delays, etc).
Installation and upgrading options
1 Why are there 2 .dll/vst files installed with the plugin?
Both versions of the .dll are optimized for different tasks, as both versions of the plugins have different internal latency settings. The Nebula 'normal' is best for dynamic FXs as compressors, preamps, eqs, mics, etc. The Nebula 'reverb' is best for space and time-variants FXs as reverbs, flangers, choruses, phaser, etc.
2 I tried to install the last version of Nebula 2 Free but it didn't work. The installer halted with an error message about invalid drive C.
Please rename your first active partition of your first hard drive as C during Nebula installation.
Yes you can. Simply create a different Nebulatemprepository folder for each installation. It is recommended that you only use the commercial version, as all the library content is included with it.
First run and general usability questions
1 Why is Nebula a "CPU killer" or so resource hungry?
1) You didn't download and install any emulation presets. Please head to the Acustica Audio website, download and install Nebula libraries then reload Nebula in your audio VST host.
or, if you have downloaded and installed the emulation presets in a different folder to your Nebula folder...
2) Nebula is searching for emulation presets in the wrong folder. If you navigate to your VST plugin folder you will find a folder named "Acustica", inside this folder you will see two .xml files. Those file are Nebula MAST page settings. Right click on the file, and click on edit. You should see the below text:
As you can see, the emulation presets (n2p and n2v files, stored in programs and vectors respectively) are in C:\nebulatemprepository\ (PC). On a Mac, it is located in HD/library/audio/presets/acustica audio/nebula. Check if Nebula is searching for your emulation preset in the right folder on your hard drive by referring to this file.
The commercial library of Nebula2 will be compatible with Nebula2 free. The commercial library of Nebula3 will be compatible with Nebula2. The other free releases of Nebula3 (free and LE) are compatible with Nebula3 library.
Nebula and Nebula Reverb are exactly the same. You could compare the dll even with a hex editor, you'll find that they are exactly the same file. The difference is the XML file. In Nebula reverb, there is a large latency in order to reduce the CPU load. For this reason it is configured for accessing longer kernels, which in general yields better quality. If you can, it's better to always use the reverb version. However, keep in mind that this version adds 16K latency. As Cubase 4 and other DAW hosts support a variable delay compensation, we will most likely move to a single plugin that includes variable latency and performance settings. At the moment for users with Cubase SX2, Logic 6 and other old hosts (that do not support variable latency), introducing a real-time delay-compensating version would only cause confusion.
You need to send us your *.SER file which is created after you have run Nebula for first time. The *.SER file will appear in your nebulatempdirectory, a directory that you would have specified when installing Nebula. You will need to install the key that we send you. The installer for the key will prompt you to install the keys to the nebulatemprepository.
Please check your Nebulatemprespository folder on your harddrive. Within this folder you will find a programs folder and vectors folder. Copy the presets you have downloaded from Acustica Audio website into these folders. Programs are files that have the *.n2p extension and should be copied into the programs folder. Vector files have the *.n2v extension and should be copied into the vector folder.
Yes - Nebula2 and Nebula3 have the same engine.
When upgrading, you installed a setup version instead of an upgrade version. As the setup version overwrites your current user key, you will need to send your SER file to firstname.lastname@example.org and we will provide you with a new key.
1 What is liquidity?
Liquidity is an interesting concept, and to understand it requires some understanding of the technology behind Nebula. Simply speaking, liquidity is a parameter that controls the way in which impulse responses are mixed together in the time domain. It is implemented via smooth technology created by Acustica Audio. The liquidity fader controls the fades between these impulse responses. High liquidity levels can bring out phase issues and too low liquidity levels can create little clicks or noise on some presets emulations. The slider is present so you can adjust the amount of liquidity to match the desired result. Typically, keeping the liquidity level low will result in a more realistic emulation of sampled equipment.
You have to save the setting regarding the colour scheme and MAST page parameters before they are recalled. So, first change the colour scheme, and then click the save button.
In this case the attack and release sliders represent values concerning the operation of the internals of the Nebula engine, specifically the way in which it switches between impulse response samples or "snapshots" of the kernels it is processing. For example, if you have an attack of 1ms on a preamp it switches kernels at a 1ms slew rate. Think of these parameters as being the controls of an envelope follower. It is a good idea to keep these values low, as doing so will make Nebula emulate sampled equipment with a behaviour that is closest to that of the original equipment.
No. This only depends on the value of the DSP BUFFER in the MAST page. You are able to see the latency value in the DISK page.
If you hold the shift key, you will be able to make smaller, finer adjustments on the sliders.
You should look into upgrading your product from Nebula2 to Nebula3. Nebula2 has no labelled faders.
Yes. We will release programs based on parallel compression soon.
Nebula (The Dynamic FX version: preamp, eqs, compressors, etc. Not the reverb version!) is real time capable from version 1.2.715. If you set the AHEAD parameter to zero and DSPBUFFER to 16 in the MAST page, you can achieve a latency of 20 samples (that's very close to "zero" latency).
The trouble with ReaMote or other similar application is, you need a working DAW with plugins on the remote desktop. A custom solution would be much better. Think of a gaming PC with an nvidia card. You connect to your laptop for a while and complete the mix, without installing anything else than Nebula Server.
Nebula Reverb version has around 16K of latency, which is not so much, around 1/3 of second. Some hosts behave strangely when latency is more than 8192. If you are having problems with the default Nebula Reverb latency, you can tune it - open your xml descriptor and change DSPBUFFER parameter as follows:
with this setting CPU load is doubled, but the latency is half of what it was previously.
12 My host doesn't have manual latency compensation. How can I figure out the latency of the plugin to manually compensate?
The formula is: [DSPBUFFER * (1 + THREADS)] + (rate / AHEADLENGTH) + (rate / FORCEOFFSET)
Rate value is the Sample rate frequency of the VST host (example 44100). DSPBUFFER, THREADS, AHEADLENGTH and FORCEOFFSET are parameters that you can find in your .xml descriptor.
For the standard version cammon values are:
At this time, unfortunatly no. However, there is a fadeout option that can be used to modify the tail. To sample all aspects and parameters of a digital reverb would take a very long time - at some point in future developments there may be more complex presets that utilize these settings.
Go to the MAST page and increase RATE CNV parameter to maximum and press save. Exit and reload Nebula. It should now be ok.
Some versions of VST hosts only have plugin delay compensation active on inserts. Others, like Logic 5.2 can compensate plugin delay for a small fixed amount of latency time. This is due to the fact that Nebula Reverb plugin has a big buffer that Logic 5.2 cannot compensate.
Now save the file and test your settings. Things should now work much more smoothly.
Ableton Live, Adobe Audition, Apple Logic*, Cakewalk Sonar, Cakewalk Guitar Tracks Pro, Cockos Reaper, Digidesing Pro Tools HD (thru FXpansion VST to RTAS adapter), Digidesing Pro Tools LE** (thru FXpansion VST to RTAS adapter), Digidesing Pro Tools M-powered** (thru FXpansion VST to RTAS adapter), FASoft n-Track, Hermann Seib SAVIhost, Image Line FL Studio*, Jørgen Aase EnergyXT, Merging Technologies Pyramix, Mackie Tracktion, Magix Sequoia, Magix Samplitude, Native Instruments Kore, Steinberg Cubasis*, Steinberg Cubase VST 32*, Steinberg Cubase SX*, Steinberg Nuendo, Steinberg V-Stack, Steinberg WaveLab*, Sony ACID, Sony Sound Forge, Sony Vegas, Synapse Audio Orion, Tascam GigaStudio, RML Labs SAW, Wavosaur and Xlutop chainer.
* Strange behavior was reported on latency compensation.
In order to use Nebula or other VST plugin software from Acustica Audio you should install the VST to RTAS Adapter from FXpansion.
Information about FXpansion VST-RTAS Adapter:
The adapter supports all the latest VST2 features, such as sample accurate timing, parameter automation, automatic tempo synchronization, easy preset browsing, and patch/preset interchange with other VST hosts such as Steinberg's Cubase VST, Cubase SX and Nuendo series.
For Pro tools 7.x, including LE, TDM/HD and M-Powered -> FXpansion VST to RTAS Adapter version 2.
Unlike Digidesign's Pro tools HD systems, Pro tools LE and Pro tools M-powered do not have automatic delay compensation. In order to compensate for delay, you should use Digidesign's digitrack time adjuster.
The MAST page provides information about the global Nebula settings. Displayed immediately below the status bar is the Nebula version number and release details.
A – METERS
B – OPT TIMED
C – OPT FREQD
D – PROG RATE
E – QUALITY
F – SCAN BOOT
G – DSP BUFF
H – COLOUR
I – BOOT
J – LOADER PR
K – OFFSET
L – L TIMED
M – L FREQD
N – GUI RATE
O – KERNELS
P – AHEAD
Q – RATE CNV
R – GHOST MODE
S – AUTOTHREAD
In the upper-right corner there a couple of buttons:
A – ABOUT
B – SAVE
No. Internal routing including M/S processing and audio sidechain will be implemented in the surround version which doesn't exist yet, but is planned for the future.
Nebula reverb version has around 16000 samples of delay. If you have a powerful PC you could reduce the delay to around 8000 samples (which is the standard delay introduced by the mastering suites) but the CPU load will be twice the amount it was at 16000 samples. The CPU load and delay introduced are not bugs. This is due to the intensive and complex nature of the plugin. If you are interested in decreasing the delay generated by Nebula into your VST host, please read usage questions in the Acustica Audio F.A.Q.
Also note that the bypass will emit a slight click when engaged. We may look at integrating a smoother morph into the bypassed signal in future releases.
22 There seems to be a mismatch between Reaper and Nebula Reverb CPU usage meters. I loaded up 10 instances as sends in reaper and I got very high CPU usage (56%). I compared the same scenario in Live and got real low CPU usage (6%) and no artifacts like the Reaper.
When reverbs are loaded, ghost mode is active and the reverb tail is long, the Nebula engine opens new background threads for processing and returns immediatly to the host. This minimizes the amount of spikes - but the host can't accurately measure the CPU load in this way.
Nebula can identify when ghost mode is deactivated, and will 'sleep' before required to do further processing - in this way you will occasionally get usage spikes (as the plugin 'wakes up' to do further processing). This is especially true if you have a slow PC.
For a proper test about differences in Live and Reaper you should check differences in Windows task manager. Sometimes there is a higher CPU load simply because a sequencer is using shorter data blocks. Ensure your sequencer is using the block size selected in your ASIO settings.
For a temporary workaround disable multithreading in the Nebula Reverb XML. In your VST plugins folder you will have an Acustica folder - open Nebula3 reverb.xml and change THREADS setting from 1 to 0 (right click and select 'edit' on the file). If realtime performances becomes worse, increase your ASIO (audio) buffer size in your VST host preferences.
When downgrading, you installed a 'setup' version. As the setup version overwrites your current user key, you will need to send your SER file to email@example.com and we will provide you with a new key.
No Nebula preamps change the audio level. This means that: for an input of 1.00 dB you will have an output of 1.00 dB. If you tune Nebula meters to RMS AES 17 a sinewave of 1 Khz will be exactly -3 dB RMS and for a peak meter the same sinewave will be 0 dB, which means the maximum value in the digital domain. No gain changes will occur.
If you sample something that in the real world is not 0 dB (for example a tape that was sampled at -6 dB for the maximum value), then Nebula will reduce your gain (in this case 0dB will be mapped to -6dB). You could tune a GAINPAD function in order to map correctly the 0 dB in your inputs to 0 dB in your outputs, adding a constant 6 dB.
We did exactly use this adjustment for our "color tape" program. It means that it will saturate in the right way, but the gain will be compensated exactly like you could do with a real tape - adding +6 dB of gain after recording.
Library and emulation presets questions
1 Now that I've bought Nebula, can I sell all of my hardware gear?
While Nebula does a fantastic job of emulating high-end audio hardware, we don't want people to have the misconception that Nebula is 100% accurate. While in some cases we are very close (and make you reconsider your investments!), Nebula emulates these units and as a result is not always a spot-on reproduction of the sound.
Slider 8 is always assigned to the liquidity parameter. For more information on what this slider does, please read point 7.1 of this FAQ.
Nothing strange was detected in our testing (@ 44100 Hz SR and 48000 Hz SR with 220 Hz test tone). In testing 44100Hz we had 0 dB reduction for 1:1. In testing 48000 Hz we had -0.714dB reduction. Why is this? The explanation is simple - Samples of this original device were converted. During the conversion step the energy of these samples were not preserved correctly. The rate conversion algorithm will be more accurate in a future release.
As a short note: This is not a software based algorithm, the preset represents a piece of hardware equipment reproduced in software. Remember: Hardware may not be accurate or precise!
Yes. The Boeing 747 compressor emulation preset is puchy (like the original hardware unit), especially with fast attack/release settings. The Boeing 747 emulation preset is also punchy when compared to other compression plug-ins. It's not a matter of Acustica Audio emulations - this is due to the sound of the sampled hardware unit. The Being 747 compressor emulation preset is not 100% perfect, but there is the "feeling" of the original unit inside.
If you need less punch from the compressor, tune the lookahead parameter. Go to the MAST page and increase the lookahead parameter to 1.5 milliseconds. Hit 'save' on MAST page then go to the VIEW page and reload the Boeing 747 compressor preset emulation. Use a fast attack.
Remember: Keep the lookahead parameter around 0 milliseconds in order to keep the latency low. You can change this setting as desired.
Load the analog channel emualtion preset and then navigate to the KERN page. Increase the H1 kernel value to 10 milliseconds.
If FREQD value is reduced in the KERN page, the reverb tail will be shortened. If you need it shorter, you can also reduce the FADE value. Once you have made your changes, press SAVE and reload the emulation preset.
There are some emulation presets in Nebula that have a strange/wierd behaviour. These were attempts to sample old cassette decks and have been largely unpopular. As these were sampled from an old casette deck, there was a lot of wow and flutter which in turn are present on these emulation presets. We will revisit sampling these in better manner in the future, utilizing a better cassette deck. We have kept these programs in the preset emulation library; not just because they are good emulations, but because they are useful for creative processing as well.
8 Why do I get stepping noise in chorus programs?
This is a limitation in the technology used in Nebula. Nebula is playing different kernels like a sequence of photos in a camera. When these are played in rapid succession, it is like a "movie" of kernels at a selected frame rate (PROG RATE). The transition between these can be smoothed by the SMOOTH algorithm.
The basic SMOOTH algorithm (and the one you will use the most) is LIQUIDITY and is available in every emulation preset. LIQUIDITY is the basic algorithm and isn't the only one. The second version is called SMOOTH2, and it is available from the GLOB page. You can reach this setting from the EDIT button. Effects that are time variant in this manner have a pre-defined SMOOTH algorithm, but you are able to edit this . There are more than 20 and we are adding new ones. For more information on Liquidity refer to the previous topic in section 7.1.
Now you understand the limits. Why do the phasers sound so good? You could ask. Because in this case Nebula is synchronizing them, and in general phasers have a simple kernel (with a single peak). Choruses and flangers, especially if digital, are a lot more complex, because they have many peaks (and the peaks are moving). In this case the smooth algorithm vital to proper operation.
9 Can reverbs be tweaked with regards to predelay, length of decay, etc?
At this time, unfortunatly no. However, there is a fadeout option that can be used to modify the tail (sampling all aspects and parameters of a digital reverb would take a very long time unfortunately). At some point there may be some more complicated presets, but at this point you are limited to the basic settings.
These preamps were sampled when only one kernel was available in the Nebula Sampler (back in an older release). Now that we have 10 kernels working we are able to offer more complex emulations.
Open a less transparent compressor preset like the Yakuzi preset. Set a higher compression ratio (slider 4), and lower the threshold (slider 3), until the reduction label is showing something different than 0 dB. Your output will have a lower gain level, which you can compensate with the makeup (slider 7). Adjust attack (slider 1) and release (slider 2) times to taste. You should see a change in the levels.
A tip: when dealing with a very transparent compressor emulation preset like the Being 747, try adding some delay or reverb on the sound before the compressor and listen to what happens to the tail, this is one of the easiest ways to check that a compressor is really doing something (and also to get a feeling for what it does).
Every software compressor adds distortion but this distortion is distributed on every incoming sample, but you don't hear it. When using a preset, Nebula calculates the next kernel for every PROG RATE period. In this case, the distortion is stepped and occurs when kernels are calculated again. We are trying to minimize the changement using the liquidity parameter, or have a similar behaviour to other compressors (distributing the distortion on every incoming sample) using other SMOOTH parameter. For abrupt level switching, our technology is not perfect and can run into problems as a result. The only solution is lowering PROG RATE value which is possible with the last generation of CPU processors.
In addition, you can tune the LOOK AHEAD parameter in the MAST page. LOOK AHEAD will compensate the delay of the PROG RATE value, and prevent an extra clipping (but compressors will behave more like a limiter).
Both Nebula2 and Nebula3 share the same library and both can load the free libraries. All paid version of Nebula contain all free emulation presets from Nebula 2 Free, Nebula3 CM, Nebula3 KM, Nebula3 DR etc.
2020 Yakuzi C II compressor emulation preset is a low frecuency response enhanced version of 2020 Yakuzi compressor emulation preset.
Nebula is flexible in that a library may be stored in multiple locations on a hard drive, or across many drives.
DATAPATH: This is the main folder, it should contain a "Programs" directory, a "Vectors" directory "Setups" folder. The "Programs" directory should contain at minimum the file named "Init.n2p". The "Vectors" directory should contain at minimum a file named "Init.n2v". You can choose the drive location you prefer, for example x:\myprincipalrepository.
The "Setups" folder should have the Rom file linking programs and vector into Nebula program list. (the default name is rom.xml, as expressed in SETUPROMNAME)
ALTERNATIVEDATAPATH: you can use as many alternative data paths as you want. Each should contain a "Programs" and a "Vectors" folder. You can specify as many drives as you want (for example x:\newdatapath, z:\newdatapath, h:\newdatapath). If Nebula doesn't find a path it simply will not load those programs, but will not crash. So in this respect, these alternativedatapaths could be also removable drives.
PROGRAMSSUBPATH: is the name for ALL programs sub-folders. The default value is "Programs". Don't change it.
VECTORSSUBPATH: is the name for ALL vectors sub-folders. The default value is "Vectors". Don't change it.
16 Is it true that the different EQ bands affect each other? Is it true that Q of an EQ band will change dynamically according to the amount of gain or attenuation you add to the given frequency? Can we have emulation presets that have more than one band at the same time that affect each other?
The only way to replicate this behaviour is sampling everything at every possible combination. It will work because Nebula has no limit in how many variables you are adding to emulate the model.
Nebula does it but we didn't use extremely loud test-tone on this emulation preset - we were looking for a clean behaviour.
18 On Nebula preset emulations do I hear distortion of the sampled hardware or that of the audio converters?
We don't distort converters and we use converters that are good quality, so most of the time you are hearing sampled gear distortion. In addition some digital reverbs were sampled thru digital IO.
These versions of emulation presets are obtained forcing a neutral linear response for the flat position. There is a 1000 band equalizer which calculated one or more corrections, and applied them to the whole collection of kernels. We are speaking about neglegible corrections. Mistakes were caused by DAC converters, or by the design of the original gear. The frequency response was not perfectly flat, so not suitable for mastering purpose. In a mastering session, when you insert an eq you expect that you have not changed anything till you don't move something. Then you apply just a 0.5 dB here and there, and the job is done.
For the first time we are joining the power of hardware, with its detailed phase response and the beauty of software.
When you are outside the box, you are losing something, especially with mid-priced converters and cables and other external gear (not including clock problems).
Old plate III is a 32 bit emulation preset and Old plate IV is a 64 bits emulation preset. Yes, we are speaking about an old EMT plate reverb.
Nebula skin questions
1 Explain me how to change the skin in Nebula?
First You should download a new skin. Inside your nebulatemprepository you will find a Skins folder and a file named OverwriteMe.N2S. Just change OverwriteMe.N2S to OverwriteMe.old and the new skin xx.N2S to OverwriteMe.N2S.
There is a little application for automatic skin change developed by asseca user. For more information about this topic read this
2 What should I do in order to create my own skin for Nebula?
A tutorial about this topic is available. Please contact asseca thru a PM in the Acustica Audio forum. In addition NAT2 will encrypt skins into .N2S format soon.
V.V.K.T. and Acustica Audio technology questions
1 What does static convolution and dynamic convolution mean?
When you send a signal to a room thru a loudspeaker or to a hardware device through a cable it doesn't output the same result, it changes if you use a different amplitudes. A hardware device in general adds modulation and that means the output will be changed in time. In addtion when you send a signal to an effect unit you could get distortion and this distortion will changes the content of the incoming signal on the output. So it's impossible to make a good replica of a room reverb or a FX unit with only one impulse response sample. Standard convolution software and hardware (most of them only work as reverb FX processors) only uses one impulse response sample. Dynamic convolution hardware only uses a small amount of impulse response samples. Software and hardware that only uses one impulse response sample works with static convolution.
2 Does Nebula use dynamic convolution?
Dynamic convolution applies an impulse response for each different input level. Distortion is a consequence of the impulse response switching and memoryless, only depending on the actual value of the input assessment. Dynamic volterra kerneling applies the same group of kernels to a group of samples, morphing them in real-time. There is a direct correspondence between harmonics and kernels.
3 Acustica Audio software is based on Angelo Farina's non-linear convolution techniques? Was he involved in the developement?
Angelo Farina was not involved directly, but Giancarlo and Francesco met him a couple of times. A piece of Nebula architecture is based on Farina's technique. Angelo published an AES paper regarding Nebula and blind tests.
4 Will it be possible to have inverse impulse response kernels?
We are researching this technique. Our first approach will be working on a method that will allow samplists to compensate for the sound and LF roll-off of AD and DA converters and cables [edit: released, this technique is named S.T.O.N.E.]
5 Could I exaggerate distortion in a emulation preset?
It is already implemented. Take the "new man" mic example, you could play with those settings simply by using the sliders.
6 How does the even and odd harmonics sampling work?
It's part of the whole science behind Nebula and based on the theory of volterra series. Due to the math behind it you can seperate the odd and even harmonics with our deconvolution engine. If you have our deconvolver you don't need to worry about it, it will do it for you in a nearly one click solution, and it's not synthetically, you receive the harmonics from the recorded test tone which you can generate inside N.A.T.
7 Odd and even harmonics are added synthetically by the plugin?
No. Odd and even harmonic are captured by N.A.T. and they came from the unit that you are sampling.
8 Are there plans to combine both dll/vst version of Nebula into the one plugin?
Yes, we will have an unified plugin soon. New VST 2.4 hosts like Steinberg Cubase 4 are currently supporting a variable delay compensation. As soon as we have a working host with this feature we will implement it.
9 I have no idea if the quality of the DA and AD converters are important when it comes to the quality of Nebula emulation presets and the sampling process?
Yes. The entire sampling ring should be top quality in order to get pristine emulation presets.
Where can I exchange Nebula presets?
|There is a website set up by some Nebula enthusiasts - please visit them over at www.nebula-programs.com|
How can I run Nebula on CUDA?
We are supporting CUDA version 2.3 with the commercial version of Nebula Reverb. If you are not a customer, you need to make a request to firstname.lastname@example.org, or send a support email by using the contact form on the site. If you are a customer there is a proper thread in the customer section. As a brief you need to download and install the latest nvidia CUDAversion , nvidia CUDA toolkit and Acustica Audio's CUDA bridge. Once downloaded and installed you're ready to use with Nebula.
To use CUDA for reverbs, follow the steps below:
(You will find these files in C:\nebulatemprepository)
Copy the "Nebula3 Reverb.dll"file and rename it "Nebula3 CUDA.dll" or choose a name that you like.
Copy the "Nebula3 Reverb.xml" and rename it as above (e.g. "Nebula3 CUDA.xml")
Go to the MAST page in Nebula, set the FREQD to mode 15 (CUDA 2ch Mono).
DO NOT CHANGE OPT TIMED PARAMETER!
This version will now process reverbs on your Graphics Card, rather than your host CPU
CUDA is supported on mac OSX aswell. The installation precedure is described in the forum.
What is CUDA?
|Nvidia CUDA technology is a fundamentally new computing architecture that enables the GPU to solve complex computational problems in consumer, business, and technical applications. CUDA (Compute Unified Device Architecture) technology gives computationally intensive applications access to the tremendous processing power of NVIDIA graphics processing units (GPUs) through a revolutionary new programming interface. Providing orders of magnitude, more performance and simplifying software development by using the standard C language, CUDA technology enables developers to create innovative solutions for data-intensive problems. For advanced research and language development, CUDA includes a low level assembly language layer and driver interface.|
What is N.A.T.?
|N.A.T. is the FX Sampler. It is able to sample the great part of audio signal processors like: compressors, equalizers, filters, preamplificators, guitar and bass amplificators, tapes, reverbs, microphones, Time variant effects and HRTF (head related transfer function) N.A.T. is based on VVKT too. It permits to sample the original linear impulse responses (traditional convolution), the original not linear impulse responses (original harmonic distorsions), the dynamic behaviour (for example compressors) and the time variant behaviour (for example choruses, flangers and generic time variant processors).|