Steinberg Wavelab 6
Jun 7, 2008
Wavelab has a long and honorable history going back to the mid-90s, when I first was introduced to it. At that time, there was only one dedicated premastering hardware device that all the major mastering/ premastering houses used: a $20,000+ unit that rendered excellent results. However, it was self-contained and totally proprietary…and most of us simply couldn't afford one of our own. When I first heard about Wavelab and VST, I went to a demonstration conducted by Steinberg's Greg Ondo. I was, admittedly, very skeptical that either program could possibly deliver professional results but I was curious and open-minded. At the end of Greg's demo, I ordered both programs on the spot. Initially, VST had a few kinks but Wavelab worked almost flawlessly right out of the box.
Wavelab 6.0 installed without any problems as have all the previous versions I have used. Nothing could be simple. You load the disk in and let it run. Just follow the online prompts and, within a couple of minutes, this massive program is loaded and ready to go.
In my case, getting started was about giving the printed manual a quick look.
If you're a first-time user, be warned: the manual is a whopping 756 pages long…excluding the index! However, don't become overwhelmed!
When the program is running, all you need to do is to click on the "Help/ Operation" drop-down menu and you will access the online '.pdf' version of the manual that contains even more pages of information: 818 pages, excluding the index! But, these manuals are well written and very easy to understand.
If you're new to Wavelab, or this kind of software in general, you would probably benefit from skimming through the printed/ online manual to give you a better idea of how to "connect the dots". If you are a previous Wavelab user, however, you will recognize most of the GUIs and various drop-down menus and the new features will be very easy to master. Both manuals are clearly written and both contain helpful hints to help you start using this powerful program ASAP.
I and others have reviewed previous versions of Wavelab so I won't repeat what has already been described, except where it intertwines with the new features.
Even without skimming the manuals, the GUI was very familiar and it is largely based on the same familiar conventional recorder transport controls from previous versions. Steinberg claims that there are over 120 new features in Wavelab 6.0. I would have to write an entire book in order to cover all of them. Therefore, I will concentrate on the most notable features that I was able to test and that I found significant.
The SPECTRUM EDITOR is a very exacting and finite editor for manipulating short events within the frequency and time domain. It is important because it approaches editing waveforms in a new way that is an important addition to the conventional editing approaches. It is not meant to replace these but to give an exacting form of editing that uses very powerful filters for creative SFX.
Within the Spectrum option, you are offered a number of choices. "Style" allows you to choose between a monochrome or colorized representation. "Logarithmic frequency scale" displays each octave of the frequency-range equally spaced vertically. However, the manual suggests using the "linear mode" instead because higher frequencies are better displayed over a larger region. And, frequently, the higher frequencies contain the useful coloration you will want to manipulate.
With "Range", any dB level below what you pick will not be displayed, which allows you to concentrate on the more audible dB areas. "Audio gain for analysis" is my favorite option because it allows you to, non-destructively, boost the gain to find low-level artifacts you may want to remove. "Resolution" selects the size of the frequency analysis but be warned: the higher the value, the more frequencies will be analyzed but the results will be less accurate in the time domain. The time and frequency resolution will be displayed in the lower right part of the menu.
To access this powerful editor, click on the "Spectrum selection" icon at the top of the screen. This opens a colorized or monochrome spectrogram (your choice, via the "Spectrum Options" menu, activated by clicking on the downward pointing black arrow at the bottom right of the screen!). Personally, I prefer the colorized version because, for me, it's easier to read: it ranges from black/ purple for minimum intensity to red for maximum. The frequencies are displayed logically: the lower frequencies at the bottom of the spectrogram, the higher frequencies at the top. The vertical ruler at the left edge of the spectrogram displays the frequencies in Hz. The status bar at the lower right corner of the main screen will display the exact time and frequencies wherever you point your mouse. You can choose the resolution for these by clicking on the "Spectrum Options" menu. Clicking on the "Spectrum selection" icon also opens the editor which allows you to choose the type of processing you want and set a variety of parameters depending on your choices.
There are two primary modes you can use: "Surgical processing" and "Master Section processing". Surgical processing is done offline and is limited to no more than 10 seconds in length. It is used for audio restoration. Once you have selected the Spectrum option, as described, you, then, select the area you want to edit by dragging your pointer to select the length and width.
SURGICAL PROCESSING offers a number of options. "Damp", "Fade Out", "Fade In", "Fade out then in" and "Fade in then out" allow you to choose one of three filters: "Pass-band", "Low-pass" and "High-pass" filters. "Damp" allows you to attenuate the level of the frequencies of the selected region. "Fade Out" gradually attenuates the selected frequency-range over time as determined by the "Gain" setting in that editor: -48, -14 or 0 dB or .4, 20 or 100%, depending upon whether you choose "dB" or "Percentage". "Fade In" is the reverse of "Fade Out" and uses the same parameters. "Fade out then in" is a combination of the previous two choices: the selected frequencies fade out to the midpoint of the selected region and then fade back in. "Fade in then out" is the reverse of this and both choices use the same "Gain" settings as "Fade Out/ In".
There are, also, "Blur peaks" and "Dispersion" options. "Blur peaks" analyzes the loudest frequencies and boosts or cuts them, depending on the gain value you choose. If a negative gain value is chosen, the chosen frequencies will be sublimated within the mix and "disappear". This works best on sudden bursts of unwanted frequencies of short duration: an AC unit kicking in, or momentary feedback, for instance. "Dispersion" works best on lower frequencies by masking the unwanted frequencies without affecting the average frequency spectrum. The aforementioned three pass filters are not available with "Blur peaks" or "Dispersion".
There are also "Copy" operations available within the "Surgical Processing" mode. This allows you to copy your Source settings to another Destination. There are some restrictions in using this, however! The Source and Destination must be of equal durations and have the same frequency ranges. Source and Destination must be in the same audio file. The filter's steepness must be set to a high or infinite value. The Copy operations should not be used on low to lower-mid frequencies to avoid obvious audio interruptions within the audio track. It works best on higher, unwanted frequencies. This feature is very useful for removing a sudden quick burst of unwanted noise and narrow frequency ranges. There are further options within the Copy feature: "Copy exactly", which does exactly as the name implies and "Copy ambience", which copies an average of the source frequencies and "blurs" them, creating a more transparent result in the Destination. There are clearly written useful examples and suggestions in the .pdf Operation Manual that go into a great more detail than I am doing here.
MASTER PROCESSING is used for frequency-specific processing. Ideally, the selected region should not be longer than 1 second. The selected region can be sent to the Master section input to be processed by it (1 thru 8 FX processors, Master level and Master output mix) or directly to the Master output mix without FX processing or level changes where the selected region will be mixed with the rest of the file, or it can be bypassed, which removes the selected region from the file, while the rest of the file can be routed to the Master Section In/ Out.
In Master Processing, you have your choice of "Pass-band", "Low-band" or "High-band" filters as described in the "Surgical processing" section. Here, you, also, select the routing of the selected and non-selected sections as described above: "Bypass", "Send to Master Section input" or "Send to Master Section output".
You can, also, select "Filter steepness" (96 dB/ octave or less) and "Crossfade time" (with presets of 20 or 10 ms. or manual gradations of decreasing increments of 1 ms.).
The online Operation Manual goes into a great more detail about applications and rendering. This is clearly written so I feel no need to repeat what is already there.
The CRYSTAL RESAMPLER is a professional-quality sample-rate converter that is extremely easy to use. It consists of two controls: "Sample rate" (anywhere from 6 kHz. to 384 kHz.) and "Quality" (from "Preview" to "Ultra"), which defines the quality of the algorithm used. The strain on the CPU is much lower…and faster… in "Preview" mode, but the quality will suffer, somewhat. "Ultra" takes longer to render but produces a superior result.
There is, also, a new LOUDNESS NORMALIZER. It's pretty straightforward and it works quite well and is, decidedly, a big improvement over earlier versions of the normalizers included with Wavelab. However, frankly, I would caution the user, here. In the professional world of premastering/ mastering, normalizers are usually considered a "4-letter word" that should be avoided whenever possible. I wrote a similar comment about normalizers in my last Wavelab review. Why? Because normalizers, by their very nature take the sum of given tracks and "normalize" them. Which means that they analyze the varying dynamic levels of individual songs or an entire album and then process the material so that the loudest parts are compressed or limited and never go above the amount you set. Unfortunately, because a dumb, unfeeling computer, lacking in aesthetics, is making these decisions, the results can frequently be less than desired. To Steinberg's credit, this version offers a lot more finite control over earlier versions and I think that it might be useful on individual tunes. However, when it comes to setting the over all levels of all the tunes on an entire album, personally, I would still prefer to do it on a song-by-song basis manually, and make relative changes while using my ears to make those decisions. Here's why: on the average album, you're going to find a wide variety of dynamic content (I hope!) both, within a given tune and from song-to-song. In an ideally mixed and premastered album, the listener is going to set the volume and tone controls on their home or car stereo and not have to keep making adjustments from tune-to-tune. The louder, more aggressive tunes should sound more dynamic than the softer ballads but there should be a reasonable balance between these two extremes that allows the differences to be heard and appreciated without having the listener making constant adjustments between the louder and quieter tunes. (Imagine a quiet evening shared with your "honey". Things are getting "interesting" and, suddenly, you have to jump up and make adjustments because the volume suddenly peaked to ear-deafening levels! Or a quiet ballad comes on and the romantic mood is broken because your "honey" asks you to turn the volume up because it's a favorite tune. Need I say more?) This is where educated human ears still triumph over a "dumb" computer. One of the things that separates human from computer generated "decisions" is that the human aesthetic value perceives such changes on a human level flaws and all. A computer does not. One of the qualities that separate a truly listenable album from a barely tolerable one is the constant change in dynamics within a song and between songs. Neutralize this and the result becomes uninteresting and "ear-fatigue" can set in very quickly. The listener probably won't know quite why he or she gets bored with a particular album but this is part of the problem.So again: I think that the new Loudness Normalizer may have some value when applied to individual cuts but I would prefer to trust my ears when "normalizing" an entire album. Granted, this process is slower but I suspect that the results will be more rewarding.
The AUDIO MONTAGE section has been a standard feature in Wavelab for a number of years now. Even in its earliest versions, it has allowed the user to record, edit, arrange and play-back audio clips on multiple tracks or channels.
This version has been upgraded to include not only CD creation but DVD-A creation, as well. In this version, it can, also, be used for multimedia, mastering and radio-spot production and more. But, this version adds even more features! You can add pictures and text to a DVD-A, plus there are new fade-editing, import/ export and batch-processing options. In a word: the whole Audio Montage section has become more sophisticated than ever before. These new and updated features are explained clearly and in great detail in the Operation Manual.
Besides the aforementioned new and updated features, there are many more, including unlimited audio file size, the DIRAC engine: a high-grade time-stretch/ pitch-shift algorhythmic process, with the ability to, now, process your audio through external hardware, PAN NORMALIZER which allows you to process the L/R channels of a stereo file to the same level setting, EFFECT MORPHING, which blends two selected audio ranges gradually, via a customized envelope for sample-accurate transitions, updated "Preference" dialogs, offering more options, "K-System" support for volume metering, batch processing of Audio Montages and improved PlugIn management for Montage, allowing you to save/ load complete processor chains or to copy/ paste PlugIns between tracks or clips. MIDI commands can, now, be assigned to Wavelab functions. Wavelab can, now, be slaved to ASIO Positioning Protocols for sample-accurate synch.
There are even more new features but, at this point, I think you get the idea: Wavelab is one powerful muther !
I tested all the features I wrote about and they all worked as they were supposed to. My only reservation is that Wavelab 6.0 has yet to become 100% "Vista" certified. I don't blame Steinberg for this situation but, rather, Microsoft for its slowness in adding the necessary fixes to get a lot of major programs not only Wavelab 6.0 totally compliant. However, I was just informed that, supposedly, Microsoft will be releasing a massive new update in March, 2008, and that this will resolve a lot of the problems that still exist between "Vista" and a lot of major 3rd-party software programs.
One can hope !
|Summary and conclusion|
As usual, each new version of Wavelab is nothing short of amazing. If there is anything that I fear it is that this wonderful program is, perhaps, on the verge of becoming a little too "full-featured" and bloated. Even though I have used Wavelab professionally in my premastering business since version "1.0", I already note a number of new features that I will, likely, never use in my work. However, the full range of Montage features will probably be useful to those who are preparing DVD-As or working in the creation of radio jingles and multimedia projects which I do not.
Wavelab 6.0 positively exudes high quality throughout. The old familiar features are still there although many, including the GUI, have been enhanced or upgraded…and the newest features will certainly be a bonus to those who can make use of them. The Spectrum Editor certainly adds extremely microscopic editing capabilities to the program and many of the included FX and processors have been upgraded, as well. Plus there are a few new ones that exhibit the same high quality that the rest of the program does.
As usual, designer, Phillipe Goutier, and his team are to be congratulated on a job well-done.
|Be the first to comment on this article|
Become a member now!
It free and fast and it'd allow you to post news, ads, messages in the forums, change your language/time setting...