2
2. Intermediate
Ok, I know my way around, I can sample and use effects, and I've released
some MODs, but just how are certain things done?
-----------------------------------------------------------------------------
Effects
~~~~~~~~~~~
There are a number of effects available in the newer trackers that we
didn't discuss in Section 1. Be sure that you are familiar with all the
standard effects before you embark this next voyage of discovery. These
effects -have- to be used properly, or they can completely destroy what would
otherwise be a good track.
Panning
~~~~~~~~~
Let’s start with Stereo Panning. This is the method by which a sound
appears to come from a certain place between two speakers.
Panning is accomplished by use of command 8 (In FT2, in others substitute
for whatever command they use).
It's a simple command to use. 800 will pan the sound to the far left, and
8FF will pan far right. Values in between these pan the sound accordingly -
880 places the sound directly in the centre, 860 places it a little to the
left, 8D0 places it quite some distance to the right.
There is also a Stereo Surround feature in a few trackers. Stereo
Surround is actually far simpler than it sounds. Once everything has been
mixed, either the left waveform or the right waveform of the stereo pair will
be inverted (turned upside down). This effect gives the impression of the
sound (yes, you guessed it!) surrounding you.
Stereo Surround works best if you are positioned directly parallel to the
centre of where the two speakers are e.g.:
Left Speaker ---------+--------- Right Speaker
You
It helps if you are fairly close to the speakers as well. Increasing the
distance between the speakers increases the surround sensation.
There is an inherent problem with this method of Stereo Surround though.
It only works well if the sound being made surround consists of mainly treble
frequencies, since most of the lower frequencies get cut out. This gives the
sound a hollow feel. Of course, you can combat this by siting yourself left
of the left speaker or right of the right speaker, to reduce the surround
effect. But why would you want to?
In most of the newer trackers, panning can also be accomplished through
the use of the Instrument Parameters. There will almost certainly be a
default panning setting. If you are lucky there will also be a panning
envelope.
The default panning has a similar job to the default volume. It sets the
instrument to a particular panning position, which gets used every time the
instrument is played without a panning command.
Panning envelopes offer greater flexibility over the stereo positioning of
an instrument.
The problem with panning is that many people don’t know anything about
panning theory and how to set up their equipment. Most seem to end up using
sounds that swing wildly from left to right. This is agony to listen to!
Soft bouncing pans can be effective, but should only be used in moderation.
Virtual Sound Sources – By XRQ
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
As we all know, musicians and music technicians left mono sound a long
time ago, simply because the stereo sound sounds much better.
The first question is WHY?
In mono there is only one source of sound and, therefore, many problems
occur when one tries to put several instruments on only one speaker. It is
very difficult to distinguish between them. They practically eat each other
and do not come out like they're supposed to.
Stereo brought us two sound sources and it seemed that the problem would
be two times easier, however this is not the case. It’s not the fact that
there are two speakers, it's just that they can give us many more sources of
sound.
The second question is HOW?
The answer lies in (what I call) "virtual sound sources" that are created
in stereo sound. Everyone who has ever listened to music notices that some
instruments come from far left (e.g. guitars), some from approximately centre
(vocals or drums) and some from the right (make up your own example). It is
described by saying that the instruments are scattered across the PANORAMA
FIELD.
Numerous experiments have shown that a man can tell apart seventeen points
in the pan-field. To hear this many he would have to have perfect hearing and
years of studio work behind him. We, the common mortals, hear only 11 or 13,
if we're lucky. These points are, in fact, my precious "virtual sound
sources", because the sound comes from there, and there, and there... But
only with two speakers!
The purpose of this writing is to accent the importance of carefully
balanced music, of a full pan-field, of a volume of every instrument in that
field which we recognise as the music.
So, the third question is - WHAT ONE SHOULD DO WITH THIS KNOWLEDGE?
Well, it would be very advisable to look on the pan-settings of your
tracker and divide the field onto as many points as you wish (not less then
seven). Well you don’t have to, I did it for you! That is, if you use
Fasttracker 2.0x. Here's the table (hex values): -
7 points
~~~~~~~~~~
00 2A 54 7F AB D5 FF
9 points
~~~~~~~~~~
00 1F 3F 5F 7F 9F BF DF FF
11 points
~~~~~~~~~~~
00 19 32 4C 65 7F 99 B2 CC E5 FF
13 points
~~~~~~~~~~~
00 15 2A 40 54 6A 7F 94 AB BE D5 E9 FF
You may have noticed that 00, 7F, and FF are always there; those are
extreme points - left, centre and right.
That's it, then. Balance your music right!
Techniques
~~~~~~~~~~~~
Echoing
~~~~~~~~~
Do you use echoes on various parts of your MODs? If not, why not? They
are an easy way of filling out the sound. Really easy to do as well. Simply
copy a channel into another empty channel, change the volume of the channel
down to under half of its current volume, and insert a row in only that
channel. Play back the pattern, if it sounds nice, you've succeeded.
Inserting only a single row will only work well at slow BPMs, however, so
keep on inserting and playing back until it sounds nice.
One point to remember, and this is something I've seen in many MODs, even
ones produced by masters (I won't give any names), is that if the echo is
fairly long a few notes will be chopped off the end of the echoed channel
when you insert rows. But these notes still exist in the original channel.
When the tune is played back the echo will appear to stop at the beginning of
each pattern, and then start again. This reduces the 'live' feel of the
entire module. Just remember to copy the chopped notes onto the beginning of
the next pattern in the playing list, and everything will sound fine.
Gating
~~~~~~~~
Another cool effect (IMHO) is gating. This is usually done with command
A. Load a long/looped sample and set it to maximum volume. Now input the
channel below (The notes can be anything, but keep the effects the same) (No
Volume Column)
C-5 1 A0F - Starts note, slides volume
--- 1 A0F - Sets volume to sample default volume, then slides volume
--- 1 A0F - Sets volume to sample default volume, then slides volume
--- 1 A0F - Sets volume to sample default volume, then slides volume
--- 1 A0F - Sets volume to sample default volume, then slides volume
--- 1 A0F - Sets volume to sample default volume, then slides volume
--- 1 A0C - Sets volume to sample default volume, then slides volume
--- 1 A08 - Sets volume to sample default volume, then slides volume
E-5 1 A0A - Starts note, slides volume
--- 1 A0A - Sets volume to sample default volume, then slides volume
--- 1 A08 - Sets volume to sample default volume, then slides volume
--- 1 A06 - Sets volume to sample default volume, then slides volume
D-5 1 A08 - Starts note, slides volume
--- 1 A08 - Sets volume to sample default volume, then slides volume
--- 1 A06 - Sets volume to sample default volume, then slides volume
--- 1 A04 - Sets volume to sample default volume, then slides volume
Now play the pattern, and you should find that you get this choppy sound
that gets less choppy with the slower slides. That choppiness is gating.
Gating works best when used on strings and vocals, but just play around and
see what you come up with.
How to Avoid Doubled Up Channels
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Doubled up channels, simply to increase an instruments volume, are extremely
bad work. Not only do they decrease the number of free channels, but
playback of the instrument will be affected. This is usually due to slight
timing errors, and can result in a muffled sound from the mix routine.
There are a couple of ways to avoid having to use doubled up channels. First
of all, the not so good ways: -
1) By physically altering the samples volume. This is possibly the worst way
of doing it. Altering the samples volume can cause both overdrive from too
much amplification, and loss of sample data when individual sample
'snapshots' reach the zero point. Repeatedly altering the volume WILL cause
these problems and result in a sample of far lower quality than was started
with.
2) By changing the default volume. This may or may not cause any
difficulties, it all depends on what tracker is being used. In one like FT2
or PT, the default volume is the same as using a volume command all the time.
To explain this, here's an example. You have an instrument that has a
default volume of 40 (hex), and you are well into composing the tune when you
decide that the instrument would sound better at 20 (hex). You change the
default volume to reflect this. But you now have a problem; all the volume
commands and slides for this sample were designed for a sample played at 40
(hex). So now the sound gets played back far too loud or disappears
occasionally, when it didn't before. It is also far more difficult to get
smooth sounding volume slides as you only have half as many volume positions
as before.
Something like Impulse Tracker overcomes this problem through the use of a
Global Volume instrument parameter. This is a relative volume level, which
means that any changes to it do not affect how commands work with it
whatsoever.
And now the really good ways: -
3) Using volume envelopes. This is my personal favourite. It works in much
the same way to the Global Volume in IT. Therefore IT users and the like can
ignore this method completely. This is how this method works.
Load a sample, and create a simple two node volume envelope that looks
something like this.
Max Volume - * Node 1
|
|
|
|
Min Volume - * Node 2
The first node should be at the top of the graph, and the second node at
the bottom. They should be as close together as possible, creating a near
vertical slope. Set Sustain on node 1.
That's it! Now, to set the samples default volume, simply slide Node 1 up
or down.
4) Halve the default volume on loading. Easy, quick, and effective.
Whenever you load a sample, change its default volume to half. This can be
done using whatever method you like (preferably through Global Volume or
method (3)). If you find when mixing that the sound needs more power you can
increase its volume, without needing to alter any other setting.
Removing the need to use doubled up channels not only improves the sound
quality and mix speed, but it makes it easier to produce a track as well.
There's less scrolling around, and you can see more of the pattern on screen
at one time.
Sampling
~~~~~~~~~~
CD Ripping
~~~~~~~~~~~~
Do you have a CD-ROM drive? If so, do you use a CD-Ripper? You should
do. A CD-Ripper will allow you to get perfect copies of audio on CDs.
You will require a CD-ROM drive and drivers which allow raw data to be
read off CDs. Below is a compatibility list that should let you know what
drives have this capability.
Drive & Interface
~~~~~~~~~~~~~~~~~~~
LG/GoldStar GCD-R540C – IDE
BTC 36x - IDE
(Ok, so it's a little incomplete at the moment!)
If your drive is listed but you seem unable to get it to read raw data,
there may be a few possible solutions.
One problem you will more than likely find in Windows 95 OSR2 and
possibly Windows 98 is that CD-Rippers will not seem to work with them. To
get around this you'll have to bypass Windows 95's 32-Bit disk drivers by
going to Control Panel/System/Performance/File-System/Troubleshooting/Disable
All 32-Bit Protect-Mode Disk Drivers. Note that you must have DOS CD-ROM
drivers installed for this to work properly.
Certain drives and set-ups will have other problems. One of which is
that the first read attempt after every reboot will fail or take forever to
start. If this happens, eject and reinsert the CD, and try reading again.
As far as I know, FT2 is the only tracker to have a ripper built in, but
it isn't very compatible. If you use DOS for tracking then a CD-ripper
called CD2Wav seems to work very well, it'll also take advantage of any 32-
Bit CD-ROM drivers installed if you run it under Windows 9x/NT. However it
can't rip specific sections of a CD. If you want a small 2 second bite of
sound from the end of the track, you have to rip everything before the part
you want, which is inconvenient and sometimes impossible.
If you want to rip CD-DA on Windows 3.1, then the only package I know of
is Digital Domain. This is quite basic, but it does the job quickly and
effectively. On Windows 9x/NT, CD-Worx would be a good choice. CD-Worx
comes in separate versions for 9x and NT, because NT uses a different way of
handling things. CD-Worx is a nice program, with features for ripping from a
variety of CD formats. Audiograbber is the one I currently use, simply
because it always seems to work, and you can specify that if any errors do
occur simply to carry on. The free version of Audiograbber does have one
slight limitation. It can only grab from a randomly selected set of half the
tracks on the CD in one session. If you want a specific track, you have to
keep on reloading it until that particular track is available!
Filtering
~~~~~~~~~~~
There are a number of features available in most good sampling programs
that can be used to improve the quality of the sound. First of all we’ll
take a look at filters, usually there will be some sort of controllable
low/high pass filter that you can use.
At their most basic form, filters are used to remove (filter) various
frequencies from the input signal. The frequencies removed may be lower than
the cut-off frequency (low pass), higher than the cut-off (high pass), in the
range between a low cut-off and a high cut-off (band pass), or outside of a
similar range (band stop).
Low pass filters give a sound a deeper, more booming
One purpose of using a low pass filter is to remove noise from a low
pitched bass sample, it can also add fatness to the sample as well. The most
important thing to remember is not to use a low pass which lets too high
frequencies through. A low pass of about 8kHz seems to work fine in removing
noise from most bass samples.
High pass filters are also useful, and can make very interesting sounds.
When used on a bass type sound, they can give it a "hollow" quality. The
higher the cut-off of the filter, the more hollow the sound.