I've had the opportunity to play about with the GMedia M-Tron VST and
have to say that I'm not very impressed. Forget about anything to do
with 'purism' here; the real problem is just how ∗shit∗ some of it
sounds.
Most of the voices are very poorly recorded indeed. I've no idea whence
these recordings were sourced, but an awful lot of them are very noisy -
far noisier than the real thing. In fact the only time I've heard
noisier Mellotrons is when they are in Streetly prior to being fixed or
serviced. Was this noise added for effect or was the machine or tapes
from which these recordings were made defective in some way?
I'd single the solo flute out for special attention here. Given the fact
that the publicise this VST as a way of playing ∗those classic sounds∗
without the need for ∗cumbersome machines and errant 1960s technology∗,
you'd think that they would get on of the most famous voices spot-on.
The trouble is that there is popping all over the middle range of the
voice. I even went to the bother of dragging frame #1 out of the closet
and doing an A/B test. No pops on the tapes, but popping all over the
VST. The conclusion is therefore that either they added this or the
tapes are buggered. Discuss.
On the subject of the flute, there is a good clue here that things are
not all they seem anyway. As far as I know (I'm guessing to some extent
here and working only on what I have heard) GMedia claim that this is a
faithful recording of a real Mellotron, note-for-note, warts and all,
etc. Hmmm. I disagree. Walking down from the top F, the most distinctive
note on the solo flute is that 'Trio note' on the top D. You know - it
sounds like a penny whistle giving a good shriek. It ain't there on the
M-Tron. Neither is the second top F either - the one that 'warbles' a
bit to itself. But if it's a verbatim account of the voice, why are they
not there? Maybe they interpolate some of the samples. Maybe they didn't
like the sound of those notes and tuned the ones below them up by a
semitone. Whatever happened, it sure isn't a faithful recording of the
whole sound.
And something else was odd about that sound too; chords sound a lot more
'musical' than on the real thing. Why is this? The obvious answer is
that it has been tuned to some extent. Not all the way - not so much
that Antares doesn't still laugh at it - but enough to make it obvious
that the voices have been tampered with.
Hmmm again. I just checked with the Gmedia web site found at
http://www.gmediamusic.com/gforce/m-tron/M-Tron.html and they say 'The
M-Tron captures all of the character of the original instrument by
sampling every note of every key and adhering to the eight second limit
of each note. This way, not only do the resultant sounds ebb and flow,
all the inherent and nostalgic magic of each sound is closely observed -
minor imperfections and all'. I disagree. Still...at least they didn;t
loop it.
Soldiering on I try out the Mk II violins. The shriller notes (∗that∗
Bb, for example, but strangely enough not ∗that∗ F#) have been excluded
here as well and are sonically almost identical (aside from pitch, of
course) to their immediate neighbours. Where the blazes did they get
this 'original instrument' from which they 'sampl[ed] every note'? I
don't know how many Mellotrons I've played but ∗none of them∗ have ever
sounded like this.
I did an A/B of the Mk II strings and got my frame bang in tune with the
M-Tron, then did a run up of the notes. Sure enough, some samples are
duller than the frame and some are definitely out of tune with it,
particularly the bottom G and the first D and E. Methinks someone has
been tuning some sounds here, and if they have done this then what else
have they done? Added a few coughs and chair scrapes? :-) Of course
tuned sets have existed for a good while now and it's possible that they
took their samples from such a set...but who made them?
And on the subject of strings...the string section. If anything is
designed to show them up, this is the one. I have the string section
here on frame #2. You all know it. Big sound. Crunching sound. As we all
know, it's a blend of cellos, violas and Mk II strings. Unfortunately,
if you play it fast enough on the M-Tron you'd be forgiven for thinking
it was nothing but cellos. The start times are ∗miles∗ off - the cellos
start -way- before the other strings. Does anyone have a string section
on their Trons that sounds like this? I sure don't. So how (again) is
this a faithful sample...unless they put it together themselves, that
is.
The choirs are hopeless. They are either muffled to the extent that you
cannot tell the male and female voices apart (oh, and the shrill female
voice on the top C has mysteriously vanished as well) or they are
brightened up so much that there is no mid or bottom range left to speak
of. They also feature a load of old Mk II voices and rhythms, none of
which - heresy! heresy! they cry - I am mad keen on and frankly none of
which I can see the average M-Tron user ever bothering with. The old
french horns, trombones and organ (among others) sound like a bunch of
sine waves subjected to slightly different filters.
And another thing that bothered me; key clicks. Some of the sounds have
very faint key clicks at the start of them. Or something. That's not
necessarily a ∗bad∗ thing (although it shows up a badly adjusted tape
set) but the timbre of the click troubled me a little. It seemed to be
strangely ∗the same∗ throughout. I recorded the sound into Sound Forge
and cut out everything but the key click, then repeated it to try and
make out what it is. It's hard to tell, but I am suspicious of it. It
∗sounds∗ like it's percussive white (or maybe pink) noise dumped at the
start of the start of the sound to mimic a click. Like I say, I might be
wrong here but it sounds very odd.
On the plus side the Mk II strings actually sound reasonably okay,
missing, botched or alterted notes notwithstanding, and the brass is
surprisingly punchy (although they have about four different brass
voices, all of which sound very similar indeed) and the Chamberlin
sounds are, though thin on the ground, very well-presented. It's also a
lot better than the existing Mellotron VSTs such as the Nanotron (which
has a flute that sounds ∗okay∗ but a string section and choir that sound
horrendous) and the bizarre Mellow-Sounds which which is the only
example I have ever seen or heard of a ten octave Mellotron with choirs
ranging from Mysteron Bass up to something only dogs can make out. Faced
with that level of competition the M-tron wins absolutely hands down. In
fact, I find it very weird that no one has ever tried to really make a
properly functioning Mellotron substitute until now, with the M-Tron and
the Memotron. If you go to a keyboard shop you'll find a myriad plastic
boxes, all of which are fitted with drawbars and come with Leslie
emulators - only because people still demand a real B3 sound. Maybe the
Mellotron market is more limited than this.
I realise that not everyone has the requisite thousands to spend on a
real Mellotron, and for that the M-Tron fills a yawning gap, but even
for the relative inexpense it would be nice to get something that does
exactly as it advertises and does not fall several yards short. It's not
impossible to sample every voice on a properly functioning Mellotron and
use that as is, without making artistic or auditory judgements on the
product that warrant editorial meddling with the sound. Maybe someone
will do this one day. GMedia certainly have not, and that's a pity.
Mike Dickson (
tron@...) M400 #996
The Official Cynic of Streetly Electronics
Streetly Sample Library
http://www.blackcat.demon.co.uk/tron/