I've noticed that other dance CDs I listen to have a fatter-sounding bass than my own productions when played on small speakers. I've got at least one song where the bass sounds fine on my Event Project Studio 6 monitors, but just does not exist on smaller speakers. Somebody suggested that I should double the bass line an octave higher, but I can't believe that that is what big-name producers do.
Reviews Editor Mike Senior replies: The main problem you've got has to do with the fact that some bass sounds don't have many high frequencies, so the moment you transfer them to small speakers, which don't reproduce low frequencies, you're stuffed! There are several solutions to this. First, at the risk of stating the obvious, choose a different bass sound. In particular, I'd recommend substituting a filtered square wave oscillator for a sine wave, if you're using analogue-style synthesis. If the new sound lacks low-end welly, then just layer a sine wave an octave below it, a technique which a large number of big-name producers use. Alternatively, if you need more attack, try looping the sustain portion of a TR808 kick drum (basically a sine wave again) and then triggering the whole sample under your bass sample — this will give you more attack and more sub-bass.
The second approach is to process your choice of bass sound differently. You probably can't just EQ the sound for more harmonics, because you can't boost things that aren't actually there, so instead you need to do things which create new high harmonics. The easiest thing is to use distortion of one sort or another. I'm not saying that you need to make your sound into a fuzz bass — just passing it hot through a tube preamp on its way to be recorded can be enough. But don't disregard guitar amp modellers, which can also work well at low (or not so low!) settings. An alternative is to try the Waves Maxxbass plug-in, which makes bass sounds more suitable for smaller speakers, although this may not create the effect you're after, because it can reduce the power of the low frequencies as well.
I have an Emu ESI32 sampler with co-axial S/PDIF I/O and am looking for a soundcard I can connect it to. I was considering ST Audio's DSP2000 until I read that it came with S/PDIF, optical and AES-EBU I/O connections as standard, which slightly confused me. I thought S/PDIF came in co-axial, optical and Toslink varieties — what is AES-EBU?
SOS Forum Post
Technical Editor Hugh Robjohns replies: S/PDIF and AES-EBU are really variations on a theme. Both are essentially high-frequency signals with a fundamental frequency of 1.5MHz and extending up into the tens of MHz, which is why they require proper RF cable, not run-of-the-mill audio cable. The basic data format is identical, although there are a few differences between the various status flags.
S/PDIF employs an unbalanced electrical signal on 75Ω cable with peak voltages of about 1V. The low voltage and unbalanced cable mean this signal is not very robust and can't typically travel distances greater than 10 metres.
The Toslink form of S/PDIF uses optical cable. Essentially, the S/PDIF digital signal is used to modulate the illumination of an LED: the opto-receiver at the destination converts the flashing light back into an electrical S/PDIF signal. The quality of the optical fibre is paramount to the robustness of the signal, and this format has the advantage that there's no galvanic connection between the source and receiving devices.
AES-EBU uses a balanced interface (110Ω) with peak voltages of up to 8V, and the signal will travel a couple of hundred meters on the right cable. While it is possible to convert between AES and S/PDIF interfaces with a simple balanced/unbalanced-wired cable, it's far better to use proper converters designed for the purpose, which are fitted with tiny pulse transformers. These are available from Canford and Studiospares.
I currently live in London and am moving to Vancouver in the summer. I'll want to take all my studio equipment with me, but are there any issues with different voltages? Will a simple adaptor suffice, or are there other things I need to consider?
SOS Forum Post
Technical Editor Hugh Robjohns replies: I believe the Canadian power supply is 110V like the USA, whereas we use 230V in the UK. The mains plugs are also different, of course, and you may be able to buy replacement plugs to fit to your mains cables or buy replacement IEC leads for the equipment fitted with IEC mains sockets. Do not use your existing mains distribution boards as they won't be rated to handle the much higher currents required on the lower-voltage mains supply.
You'll also need to check all your mains-powered gear, which will fall into one of the categories below.
Switchable: some equipment will have a switch on the rear panel to select the mains input voltage. In this case all you have to do is change the fuse and switch over to the new supply voltage when in Canada — before plugging it in. You need to change the mains fuse in the equipment itself and possibly in the mains plug, because, if the supply voltage halves, the current demand doubles to provide the same power.
SMPS: a great deal of equipment now contains 'switched mode power supplies', which are also known as 'universal' supplies. The info label on the back of the equipment near the mains inlet will say something like 100-250V, and in this case, all you will have to do is change the mains inlet fuse.
Wall warts: these may be SMPS designs, but are more usually just a simple fixed-voltage transformer. In the case of the latter, you'll have to buy replacements in Canada to provide the same output voltage/current from a 110V supply.
Finally, check the handbooks for your equipment to establish the kind of power supply and appropriate fuse ratings, and when you're in Canada, talk to a good local music shop to source appropriate alternative supplies and cables. If in doubt, consult a qualified electrician — remember, electricity can kill, even the weedy stuff they have over there! And if it doesn't get you, it might well get your gear, so be very careful and check your modifications before plugging in and switching on.
Over the next few months, I hope to be recording a trombone quartet, a wind quintet and a string quartet. At the moment I only have a pair of Rode NT1s and a pair of NT5s, so I was thinking of using the NT5s as a coincident pair quite close to the group with the NT1s further back as a spaced pair to add some depth to the recording. However, I believe omni mics would be better suited to this — can you recommend anything? I've only got around £600 to £800 and don't really fancy the Rode NT2s.
SOS Forum Post
Technical Editor Hugh Robjohns replies: I think you're on the right lines, using the NT5s either as a crossed pair or a near-coincident pair, which would give more 'spaciousness' but with slightly less precise imaging. While choosing one of these arrangements comes down to personal taste, I'd probably start with the coincident pair if you're planning to add the outriggers as well.
The outriggers don't need to be omnis, but the advantage of omnis is that they tend to give a better bass extension than cardioids, although this will be of little consequence in the case of your quartets. I would place them to add the required amount of room ambience and spaciousness. You can orientate the mics to face toward the quartet, or away or to the sides, as required to optimise the room sound. There should be no need to buy new mics, just use your ears and the ones you already have — it should sound great!
I'm having problems with the amount of noise my computer makes and I've been told to either put it outside in another room or build a box around it and fill it with deadening foam. My computer table has a small box cabinet that the computer tower sits in, which has about a three-inch clearance all the way around the computer, so just to experiment I went to a foam store and spent about two euros buying three-inch scraps, just enough to pack the machine in tight. Included in my purchase was a piece of five-inch-thick foam that I've secured in front of the machine, and I've left the back open for obvious reasons. But guess what? It's made virtually no difference whatsoever!
SOS Forum Post
Reviews Editor Mike Senior replies: I'd dust off your carpentry skills and make a box out of wood, lined internally with acoustic foam, which will probably have to be quite thin. Make the rear of it vented to allow ventilation, but make the vent S-shaped (as shown in the cutout in the PMC DB1 monitors review in the January 2003 issue) and line it with foam also. Put a door on the front to allow access to the drives etc, and keep this closed while recording.
You might find that you need to have an air vent at the front of the box as well, in which case, I'd put it in the top and put foam over it. Most of the noise of a computer is higher frequencies, so the box should give you useful attenuation just like that. I'd also be tempted to rest the computer case on neoprene rubber (also in the Studiospares catalogue) to stop vibrations from being coupled to the desk, which will act as a sounding board and amplify them.
I have some mono material that I'd like to turn into stereo — is it possible to do this in a way that sounds vaguely realistic?
SOS Forum Post
Technical Editor Hugh Robjohns replies: The basic approach to creating a useful and reliable stereo effect from a mono source that remains mono-compatible is based on using Middle and Sides (M&S) techniques. Consider the mono source as the M signal, to which we need to add a 'fake' S signal. The resulting pair will then need to be decoded back into left and right for mixing into the stereo track, of course. The advantage is that the S signal disappears in mono, so the original source track remains undamaged for the mono listener, and hopefully usefully enhanced for the stereo listener.
At the most simplistic level, the primary difference between the M and S signals is time. Imagine a source in front of an M&S mic: the direct signal reaches the M mic, while the same signal has to travel to and bounce back from a wall to reach the sideways-facing S mic. Consequently, all we need to create a useful fake S signal is delay the original by a small amount: a simple mono delay line is all that's required.
Take the source, split it in two and route one to the centre of the stereo output, and the other to a delay line. In practice, I like to pass it through a high-pass filter first and roll off the bass below about 150Hz, as I find that produces a better effect; it can become bass-heavy on the left channel if you don't do this, depending on the source material.
The secondary M signal is routed through a high-pass filter and into a delay line. The minimum delay I would use is about 7ms, extending up to about 25ms, but feel free to experiment. With classical music, you could go as high as 80ms to create a kind of symphonic hall effect. Changing the length of the delay changes the character of the stereo effect. The output of the delay line is the fake S signal, and this now needs to be combined with the M signal to decode back into stereo.
Split the output of the delay line into two signals and phase-invert one of them. Route the normal output to the left channel mix buss, and the inverted one to the right channel mix buss. Ensure that the amount of +S is exactly the same as the amount of -S, so that when summing left and right channels in mono you cancel the S signal completely: (S+(-S))=0.
The easiest way to do this is to route the two S signals through a stereo fader, which also allows you to control the apparent width of the stereo effect — but don't go mad. The result can sound impressive, but it gets tiring quickly, and the mono and stereo mixes will sound radically different if you add too much.
Hope that helps. This technique is used very widely in broadcast post-production and in mastering when 'up-converting' archive mono to stereo, or stereo to surround.
I've just bought a Phatboy v2 MIDI controller and was hoping to use it for manipulating my sequenced samples live, although it's not quite working as I'd hoped. The Phatboy is connected to an Atari STe, which is in turn connected to an Akai S2000 and a Roland MC303 (through the Akai).
MIDI Thru is active on the Atari and the MIDI I/O bars indicate that every time a knob is twisted on the Phatboy, the message is being received and passed on to my sampler. While my sampler indicates that it's receiving the controller data from the computer and Phatboy, there's no actual change in the sound. This is the same in any of the Phatboy's modes, except for two knobs that both seem to adjust volume. Can you help?
Reviews Editor Mike Senior replies: It's very rare for any MIDI controller to work as you want it to straight out of the box, unless the manufacturer has made a specific user template for the destination gear. In the case of the Akai, the modulation routing options are quite flexible, so the developers of the Phatboy couldn't possibly guess what MIDI control changes you've set to control what parameters.
My suggestion would be to use the Phatboy in Mode III, where it sends out Controllers 1 to 13. In this mode, a normal sampler program set to the same channel as the Phatboy should offer vibrato on knob one (sending MIDI Continuous Controller number 1), because the default Akai sampler program routes this Controller number to control the amount the sample pitch is affected by the first internal LFO. (I'm assuming your S2000 shares a similar setup to my S2800 here, but I might be wrong on this). MIDI Continuous Controller number 7 can be automatically linked to volume control, and often number 11 a well, so these might be the two 'volume' controls you mentioned.
My suggestion is to repurpose the relevant sampler program so that MIDI Continuous Controller number 1 controls filter cutoff, rather than pitch modulation. Your sampler manual will tell you how to do this in Edit Program mode on the LFO and Filter pages. Once you've got the hang of it, you can try setting up panning or whatever. The only restriction is that my S2800 doesn't allow me to adjust resonance easily in real time, so I'm guessing your S2000 doesn't either.
For controlling the Roland, I'd try Mode IV, which will probably allow you to control the main parameters straight away. If it doesn't work, then there's no substitute for reading the MIDI implementations and surfing through the Roland's System settings to see if things can be reassigned.
The Phatboy is fine as far as it goes, but it sounds like it's not flexible enough for you, really. If you're unable to get what you want out of it, then why not upgrade to a Kenton Control Freak, which will let you program the MIDI messages you need (including SysEx strings if necessary) over the ranges you need. The Doepfer Pocket Control might also be an option, although it's less convenient to program.
How is mixing typically done when a composition is written on a multitimbral synth? Would it make the most sense to record each sound on a separate channel in Cubase and separate the drums into their individual sounds, adding compression and EQ to each individual track before mixing down to stereo? Is that how it would typically be done?
Reviews Editors Mike Senior replies: In answer to your main question, I'm afraid there is no 'typical' way to mix down the output from MIDI sound modules. However, I'd say that you'll get much more mixing control by recording each part individually into Cubase and then mixing from the audio. Alternatively, get an analogue mixer and some outboard processors and mix through that. Either way you get more control.
If you're going to record to your computer one part at a time, I'd suggest investing in a recording channel processor, so that you can compress, EQ and A-D convert before reaching the computer — the analogue processors will probably give you a better overall sound than Cubase's plug-ins.