Are you stuck on the PC upgrade bandwagon?
This month, PC Notes discusses whether it might be possible to step off
it, as computer power finally starts to catch up with the needs and
aspirations of musicians.
When I first started writing for SOS, way
back in 1996, my PC featured an Intel 486DX33 (33MHz) processor, 4MB
RAM, a 250MB hard drive and a Sound Galaxy Basic 16 (16-bit only)
soundcard. It seems incredible just how far we've come with PCs and
music over the last 12 years, and how increasing processing power has
kept on moving the goalposts.
For me, the first revelation was the 'plug-in',
which arrived in late 1996 with Steinberg's Wavelab 1.5 and Sonic
Foundry's Sound Forge 4.0a (now part of the Sony Creative Software
range). Being able to add further functions to an existing application
was truly liberating, and for the first time musicians could treat an
audio signal with effect algorithms in real time within their software.
However, even a single metallic-sounding reverb
plug-in of the time could easily consume 100 percent of your available
processing power, so I upgraded to a faster PC with a Cyrix P166+
processor (clock speed 133MHz), 16MB RAM and a 1GB hard drive. I had
just climbed onto the upgrade bandwagon, and during subsequent years I
formulated two good rules of thumb, although one eventually proved less
durable than the other:
If
you want your computer to be able to run the latest audio software, you
will, on average, need a major computer upgrade every two years.When
upgrading, you should ensure that your processing power increases by at
least 50 percent, so that the expense is justified by a significant
improvement in performance.
Sure enough, two years later I upgraded to a Pentium
II 300MHz processor, 64MB RAM and a 30GB hard drive, largely so that I
could run some of the new real-time soft synths, such as Native
Instruments' Generator (later to evolve into Reaktor). This machine
lasted me another couple of years until I was desperate for more
processing power and switched to a Pentium III 1GHz processor, 256MB of
RAM and a couple of 30GB hard drives. Two years later, in late 2003, the
next upgrade became essential to cope with all the latest software, and
this time I bought a Pentium 4 2.8GHz processor, 1GB of RAM and a
couple of 80GB hard drives.
Divergence Of Demand
However, at this point, while my second rule of
thumb still held true, the first began to break down: I was happy with
this machine for three years, and it would still be perfectly adequate
had I not decided to abandon my MIDI hardware synths and move over to
entirely 'in the box' song creation with soft synths and software
samplers.
So in December 2006 I invested in an E6600 dual-core
CPU with two cores running at 2.4GHz, offering a colossal 300 percent
increase in processing power over my previous machine. Even a year later
this might seem like small fry compared with today's quad-core
processors, yet the majority of my songs have yet to use more than 50
percent of this dual-core model.
Over the last couple of years I've noted a growing
number of musicians also getting off the PC upgrade bandwagon. While
some are currently buying and building quad-core machines, others
decided some time ago that they were perfectly happy with the hardware
and software they were already running, especially if it was rock-solid
and bug-free.
Available computing power finally seems to be
starting to outstrip the needs of musicians, and many of us no longer
have to compromise when choosing plug-ins and soft synths. Moreover,
MIDI + Audio sequencers now contain so many functions some of us never
use that an increasing number of musicians are not bothering to upgrade
them any further. Will I find myself desperate to trade up my PC again
in late 2008? Somehow I very much doubt it.
VST Instruments are now the foundation of many a
musician's sonic arsenal, and are getting ever more sophisticated. To
keep pushing at the boundaries, some rely on the latest physical or
analogue modelling techniques, but with many designs it's the associated
sample library that provides the sophistication, and nowadays this
generally means many megabytes and sometimes even gigabytes of sample
data.
All you need to do to install the majority of
simpler VST plug-ins and Instruments so that they appear within your
chosen VST host application is to place the DLL file (containing the
synth code itself), plus any associated presets, in the default
'vstplugins' folder on the same C: partition as your Windows operating
system. Most larger VST Instruments also store their DLL file in the VST
plug-ins folder, but provide the option to install the associated
sample libraries on a different partition or drive. This is good news
for those who want to keep their Windows partition as small as possible,
so that backup files are quicker to create and smaller to store.
However, just recently I've noticed musicians grumbling about various
large VST Instruments that don't offer such a storage option, such as
Toontrack's EZdrummer, with 1.5GB of compressed files; Manytone Music's
ManyStation, again with a 1.5GB library; and the Yellow Tools
Independence instruments. VST Instruments created using Jeff
McClintock's SynthEdit engine also currently make you place associated
files in the same folder as the main DLL file. You can end up with many
gigabytes of data in your C: VST plug-ins folder that you'd prefer to
house elsewhere, so here are a couple of ways to slim it down.
|
Freeware Spotlight: Ferox Tape Simulator
If you want to try adding some analogue tape saturation effects to
digital sounds, you could do worse than visiting Jeroen Breebaart's web
site (www.jeroenbreebaart.com).
Jeroen has plenty of expert knowledge, having worked for seven years as
a Senior Scientist with the DSP group at Philips Research, and has
developed a very comprehensive VST tape-simulator plug-in in his spare
time. Ferox looks good and sounds good, with separate controls to tweak
Rec Level, Low gain (bass hump), High gain (HF peak), Saturation (odd
harmonic level) and Hysteresis (non-linearity). It even offers variable
Noise (although I'd personally leave that set to minimum), plus Tape
speed and Feedback controls to simulate vintage tape-echo effects. The
presets cover high-quality tape decks, cassette machines, worn-out tape
and analogue overload, and Ferox is capable of a versatile range of
quality effects, from gentle thickening to total mashing. As far as I'm
concerned, this is the king of freeware tape emulations, and gives
plenty of commercial ones a run for their money.
|
VST Slimming: Part One
The first method is to install the VST Instrument
and library to your default VST plug-ins folder, but then manually drag
all the larger sample-based files to a more convenient location (you
could create a dedicated Samples partition on a different drive for all
your libraries). Once they are in place in their new location, select
all relevant files, then right-click on the lot and select the 'Create
Shortcut' option. A batch of tiny shortcut files will appear with
identical names, and you then drag these files over to the original VST
plug-ins folder, so that the synth's DLL file will be able to link to
the real files.
This method will apparently work with some
applications, such as Propellerheads' Reason factory soundbanks, and is
also used by some commercial VST Instruments, including Spectrasonics'
Atmosphere and Trilogy, to link to their data.
That first approach won't work with SynthEdit-based
creations, or various other large VST Instruments, so here's an
alternative that's often easier to implement and should work with quite a
few VST host applications. It works because many of them let you define
multiple 'vstplugins' folders; indeed, some VST Instruments, such as
Tascam's GVI, rely on this fact; by default, GVI installs its own DLL
file inside a separate Tascam Instruments folder and expects you to
inform your sequencer of this additional VST plug-ins path.
First, choose a destination for any larger VSTi
installs without their own library destination options, and create a new
VST plug-ins folder there. I placed mine on my S:\Samples partition,
where I install my other sample libraries, but gave it the name
'_vstplugins' (starting with an underscore) so it appears at the top of
the list of folders when alphabetically sorted.
Many sequencers (including Cubase 4, shown here)
allow multiple plug-in paths to be defined, so by adding another path,
such as S:\_vstplugins, to a convenient location, you can install large
VSTi libraries anywhere you wish.
|
Some sequencers, including Ableton Live and Cubase
LE, can only recognise one VST plug-ins folder, so this trick won't work
for them, but others are more helpful. Here's how to add extra VST
plug-ins folders to the list of recognised paths for a selection of
popular hosts:
Cakewalk Sonar: Use the Add button of the VST Configuration Wizard in the Tools menu.Cockos Reaper: Launch
the Preferences dialogue from the Options menu, navigate to the VST
Plug-ins area, then use the Add button adjacent to the 'VST plug-in
paths' read-out.
* Steinberg Cubase SX and Nuendo: Launch the Plug-in Information window from the Devices menu and use its Add button.
Steinberg Cubase 4:
Launch the Plug-in Information window, click on 'VST 2.x Plug-in
Paths', then click on the Add button on the new window that appears.Sony Acid Pro: Offers two VSTi Search Folder paths, which you can define from the VST Instruments tab in the Preferences dialogue.
Once your VST host application has been informed of
the extra VST plug-ins folder path, all you need to do is install the
entire VST Instrument (DLL file, sample banks, preset files, and so on)
in the new VST plug-ins folder and your sequencer will find it all
automatically. If you already have bulky VST Instruments stored on your
C: partition, you may be able to drag them across to the new
destination, but some with serial number or challenge/response copy
protection may need un-installing and re-installing in the new folder to
register their new locations correctly.
Even so, the migrations aren't likely to require
much effort. I've installed VST Instrument libraries in my S:\Samples
partition wherever possible, but I still had nearly 3GB of VST
Instruments plus sample data that I couldn't do that with, which needed
moving. I managed to transfer it into my new VST plug-ins folder in
under an hour, reducing the size of my Windows music partition from 8GB
to 5GB. The music partition image backup file now compresses to under
3GB, takes just four minutes to create and fits on a single DVD-R disk.
After a major reshuffle and 'data shrink' like this, it's wise to defragment your partition to optimise performance, and you may also want to reduce the partition's size to allow future expansion for other partitions you might have on that drive.
No comments:
Post a Comment