I've read that a microphone you would use to capture a male rock-vocal performance would produce inferior results when used to record a female vocal performance. So if I want to develop a recording sound for a female vocalist in a heavy/nu metal style, should I buy the Rode NTK, or will this be too polite and not give the right sonic qualities? Can I use a compressor, like the Joemeek VCQ6 for example, or a limiter to give the impression of power? How about the use of a preamp that has distortion?
Editor Paul White replies: When it comes to vocal recordings, a lot of people seem to look for problems where they don't exist, or where the problems are due to causes other than the choice of equipment. There are also plenty of misleading myths around that can cause you to spend money you don't need to. A good large-diaphragm capacitor mic stuck a few inches in front of any singer, male or female, should be able to capture a performance that can subsequently be tailored to a particular style using EQ and compression, either from hardware outboard gear or plug-ins. All you need is a pop shield and room acoustics that are not too intrusive. And of course a damned good singer!
If you're in the market for a new mic, a tube model such as the Rode NTK will probably provide you with a more solid rock sound than the NT1 (I'm a big NTK fan), though I have had superb results using an NT1 with both male and female performers. I think it would be fair to say that the style of compression used will affect the sound more than the difference between these mics, and for rock work, it's not uncommon to use high ratio settings (up to 10:1 or so) where the compressor is acting as something halfway between a standard compressor and a limiter. Adjust the threshold level while watching the gain-reduction meters and try to arrange it so that normal singing levels are just showing a dB or two of compression while loud phrases are 'stamped on' hard. By forcing the compressor into heavy gain-reduction during loud passages, you will be using it more as an effect than simply a gain reduction tool, but the tonal quality depends to a great extent on the choice of compressor used. Hard-knee models sometimes give a harder rock sound, but don't rule out the Joemeek optical designs as these also have a nice punchy character. If you're compressing while recording, it may be wise to err on the side of under-compression and then try something more extreme when mixing. For example, use moderate soft-knee compression when recording and hard-knee, high-ratio compression when mixing.
Many classic rock tracks were recorded to analogue tape, so using a device or plug-in that can provide or emulate tube saturation can help give the sound an edge, and use of the 'air band' EQ (broad-band boost at around 14kHz) will give the sound bite and definition without making it sound unpleasantly aggressive.
Once you have the basic sound down, you can add reverb and delay as necessary, but listen carefully to the records you wish to emulate to see what type of reverb they use. In many cases, it will be something bright and not too long or washy, so try ambience and small plate settings along with your other favourite treatments. Also be aware that the quality of the reverb unit makes a big difference, and I think most people would agree that Lexicon and TC are at the top of that particular tree at the moment.
I've been using a PC for music for years, and have dreamed of being able to afford a Mac. I've now come to the point when I'm about to buy a new computer. I've been offered a second-hand G4 (833MHz) for around the same price that I can build a 1.6GHz PC. What should I go for? I read recently that with the speed that PC development is happening it may no longer be the case that the Mac is the better option. Is this true? Which is the better platform to use?
Assistant Editor Mark Wherry replies: Trying to answer the Mac or Windows question in an objective way is a bit like putting your head in the lion's mouth, whilst standing in a snake pit, with zealous supporters of both platforms poised to fire poisonous darts in your direction. However, as a fan of both platforms, here are my views on the decision.
The first time I was truly shocked by how much ground the Mac had lost in terms of performance was when I read the July 2001 issue of the German Keyboards magazine. Rainer Hain had carried out some 'real world' musical tests, comparing the performance of the range of Macs available at the time and some Athlon-based Windows machines. When running Native Instrument's Pro 52 (which is optimised for modern processors), the 733MHz PowerMac managed 36 voices and the 850MHz Athlon 45 voices. The gap widened when running the Waldorf's PPG virtual instrument, with the Mac producing 65 voices and the Athlon giving 81 voices. However, when running instances of Waves' TrueVerb (which isn't optimised for the latest processors), the Mac managed 9 instances, compared to the 8 from the Athlon.
While these numbers aren't staggering, let's consider the pricing of these computers, which were both running Echo Mona cards at a latency of 4ms. In July 2001 the 733MHz Mac cost 4754.50 euros, while the 850MHz Athlon cost 1328.85 euros. Quite a difference.
So what's the situation like right now? Well, in the June 2002 issue of Keyboards, Rainer tested some Athlon XP 1800 (1.54GHz) and 2GHz Pentium IV machines. The 2GHz Pentium IV managed 10 TrueVerb instances, 27 Reverb 32s (the Cubase VST/32 reverb), 117 Pro 52 voices and 157 PPG voices. The Athlon XP 1800 ran 12 TrueVerbs, 25 Reverb 32s, 148 Pro 52 voices and a staggering 173 PPG voices. Unfortunately, I haven't been able to find the results for the latest crop of PowerMacs, although I've tested the new top-of-the-line 800MHz iMac I have sat in front of me, which was capable of 8 Reverb 32s, 37 Pro 52 voices, and 41 PPG voices.
Now, of course, PowerMacs have a different architecture to the iMacs, and dual-1GHz PowerMacs are incredibly powerful. I should also mention that the Windows-based computers Rainer tested were running RME cards with a latency of 1.5ms. My iMac was running with its internal hardware at 23.2ms, which obviously placed more demands on the processor. However, let's again consider the prices: a properly configured Pentium-based 'Music PC' will cost around £1500 in the UK, and I recently built an Athlon XP 1700-based computer for around £500 (without a monitor or soundcard). On the Mac side, using prices from the Apple Store as a guide, my 800MHz iMac (with its built-in monitor) costs £1938, with the cheapest G4 iMac available for £1468. Power Macs cost between £1586, for an 833MHz model without a monitor, and £3035, for a dual-1GHz model without a monitor.
So the irony of what Apple calls the megahertz myth is that while balanced system performance is crucial in the design of any computer, a £1500 Mac is nowhere near as powerful as a £1500 'Music PC'.
However, performance isn't the only issue to consider when purchasing a computer: ease of use, stability and reliability are equally important factors. In these categories, few would dispute that the Mac has a definite advantage, mainly because it's one unit designed by a single manufacturer who controls every design aspect of the hardware and software. This isn't to say that Windows-based machines can't achieve this level of comfort -- they can -- but on the whole it's usually easier to troubleshoot a Mac if you're not an experienced computer user. And it's important to remember that if you mostly record audio tracks (with hardware MIDI tracks and maybe a few plug-ins), the Mac is likely to be more than powerful enough. The number of audio tracks you can work with is usually dictated by the speed of your drives, and the interface used to carry the data. Although this requires some processor overhead, it's negligible, and I've seen incredibly old computers like a 400MHz Pentium II run over 40 audio tracks, which was barely scratching the surface of the available resources.
It's debatable that a thorn in the side of the Mac right now is the operating system. Mac OS 9 doesn't take full advantage of the latest technology, such as the performance gains possible by using multiple processors, and none of the major music software companies are shipping OS X versions of their software. Of course, OS X native versions of Cubase, Logic,Nuendo and Reason will soon be joining programs like Melodyne and Live, which are already available. However, recent OS X versions of mainstream non-musical applications have been greeted by negative efficiency reports, compared with the OS 9 version running on the same computer. Although, whether this will be the case with OS X music software is, of course, purely speculative.
While I'd never suggest Windows is perfect, Microsoft have a habit of persevering until they get it right, which is arguably what they've done with Windows XP. And, of course, Windows XP is evolutionary (as opposed to the revolutionary nature of OS X) so it can run most of your existing programs without resorting to loading a second, 'classic' OS.
So what's the bottom line? If you run vast numbers of virtual instruments and effects, want the fastest computer, change for the bus ride home, and don't mind giving up a degree of elegance, buy a Windows-based machine. And if you want the ultimate performance for the best possible price, build your own computer with an Athlon XP processor. However, if the thought of using a computer for music makes you wary, you have a reasonable budget, your music isn't solely based on virtual instruments, and you want guaranteed elegance, buy a Mac.
Assistant Editor Sam Inglis adds: While the statistics are hard to argue with, I think it's important not to underestimate the Mac's current status as industry standard. Modern Windows PCs do indeed offer much better value for money and performance, and are consequently very popular for home recording setups. However, of the many commercial and private studios I've been to in the course of interviewing artists, producers and engineers, I can't remember a single one that was PC-based. This doesn't make Macs any more powerful or affordable, but in my view, it does provide other reasons why you might choose a Mac system on wider grounds than number of CPU cycles per pound.
No self-respecting commercial recording facility can afford not to offer the option to work with Pro Tools or Logic sessions on the Mac, so if you're ever likely to want to take your own recordings into such a studio for mixing or post-production, you'll have no difficulty doing so if you're running such a system at home. And should you ever find yourself in a position to hire a room in a commercial studio, or to seek work as a programmer or engineer, you'll find knowledge of these Mac systems invaluable; any PC-specific skills you have are likely to be less relevant.
When you're setting up a home recording system on a budget, the Windows package of raw power and affordability is very attractive. However, you don't have to move that much further into the realm of semi-pro or project studios to find out why the Mac might be a better bet. The Mac's perceived superiority in this environment has been hard-won over a number of years, and although Apple seem to be doing their best to throw it away, it isn't likely to change overnight. Macs have traditionally offered greater stability, a much higher degree of uniformity (ensuring that what works in one studio will work in another), and a better range of professional recording hardware and software. Windows PCs are catching up and perhaps overtaking Macs on this last point, but have yet to sweep Apple machines out of pro studios. Pro Tools TDM systems have been available under Windows for a couple of years now, for instance, but you try to find anyone who's actually using one.
When playing my Virus keyboard through my soundcard's inputs on the computer, I seem to get a flange effect. I can only assume this is produced by MIDI notes drifting out of time, although everything does sound as if it's on the beat. Can you offer any advice?
Martin Walker replies: A flange effect is created by delaying a sound slightly and then mixing it with the original signal, so if you're getting a flange effect it's either because your sounds are passing through a software effect that you're not aware of (check your Cubase effect racks and insert slots), or that you're somehow triggering them twice in quick succession.
However, there's a much more likely solution in the case of your Virus synth: its MIDI Local mode is set to on. This is the default setting, and the one that you should use when playing any synth keyboard in stand-alone mode, so that the MIDI signals from its keyboard are directly connected to the synth. This ensures that you immediately hear sounds when you press a key.
To record your key presses in a sequencer like Cubase, you need to connect the Virus' MIDI Out to the MIDI In of your soundcard. You'll nearly always want the sequencer to play back your MIDI recordings through the synth automatically each time you start playback as well. To do this, the MIDI Out from the sequencer should be connected to the MIDI In of your synth.
Once you get more than one synth, most sequencers provide an optional software Thru function, so that whichever MIDI track is selected, when you press a key on your external keyboard it forwards the MIDI data to play that synth in real time. This makes perfect sense, except in the case of the synth already connected via local on to your keyboard. In this special case, the synth is then played directly from its Local On connection, and then again by the sequencer, after a delay determined by the MIDI interface.
Having explained the cause in some detail, the solution is actually quite simple: find the Local mode function in your Virus keyboard, and turn it off. This will break the loop, and you only need to re-enable it when not connected to a sequencer with MIDI Thru.
I recently bought Steinberg's Virtual Guitarist VST Instrument to run with Logic on my Mac, but when I try to use the plug-in, the menu to choose the guitar player is empty, telling me that the files cannot be found and to reinstall the program. No matter how many times I reinstall, I keep getting the same problem.
Mark Wherry replies: Many Mac Logic users have reported this problem when trying to run Virtual Guitarist -- it doesn't seem to be an issue for Windows users. There are many solutions, including a small program you can download from Steinberg to show both Virtual Guitarist plug-ins (Acoustic and Electric) where the data files are stored, and a less than elegant 'getting your hands dirty with ResEdit' proposal. An updated Mac installer is also being prepared, which was in beta when we went to press.
I've heard that even if the dynamic range of the signals you're recording is less than 96dB, it's worth recording at 24-bit resolution because it gives you more headroom to process signals without clipping or adding quantisation artifacts. Is this true?
SOS Forum post
Assistant Editor Sam Inglis replies: The short answer is that this isn't as much of an issue as it's often made out to be. All modern audio processing software and hardware runs at internal bit depths greater than 16: for instance, Pro Tools TDM systems use 24-bit internal processing, while other applications use 32-bit or even 48-bit algorithms. This means that 16-bit audio recordings are no more likely to lose information through rounding errors in the processing stage than 24-bit ones. Information can, however, be lost when the signal from your digital mixer, processor or software application is output at 16-bit, so it is possible to lose information by having several digital processors hooked up in series via 16-bit converters or digital links, or by repeatedly resampling a piece of audio (ie. processing it, saving it at 16-bit, processing it again, and so on). The problem in these situations is not due to making the original recording at 16-bit, but arises because of repeated truncation to 16-bit in the mixing or processing stage. If you do all your mixing and processing within one software application or hardware unit, signals that start off at 16-bit should suffer no worse than 24-bit recordings.
It is, however, true that combining several 16-bit signals can produce a sum that can't be represented by a 16-bit value, which means that there may be some point to recording the output of your digital mixer or software application at 24-bit. However, most playback media are still 16-bit anyway, so this information is going to be lost sooner or later, whether you do it here or at a separate mastering stage. The point of dither is to minimise the audible effects of this loss of information.