First Post on MIDI Board

blackscot

New member
I’ve been skulking around a couple other parts of the forum for a few weeks, learning how to make acoustic treatment to my spare-bedroom studio, and posting a couple of my recording projects (and finding them to be pretty rough compared to much of the other material being presented here). Much of my work is MIDI based though, so finally figured I ought to post some general background and questions here.

I started home recording a few years back, mostly 70’s rock classics that have especially good lead guitar parts. I don’t have the circumstances to jam with a group, so wanted to create a “virtual backup band” that I could play the lead to. As things have turned out, almost all of the backing parts are MIDI, and the lead guitar (the strat in my avatar) is usually the only audio track (other than some rhythm strumming on an acoustic guitar for a couple of tunes). So with MIDI making up really most of what I’m doing, I should probably be connecting with like-minded folks here.

I early-on got a Roland JV1010 synth, which I have been using ever since (although I upgraded my keyboard controller once along the way). At first I was recording on one of the ultra-cheapie Tascam 4-track cassette “porta studios”, so here the synth was purely an instrument with no basis in MIDI. After a year or so of becoming frustrated with the obvious limitations of this arrangement, I bought a PC along with the Digidesign Mbox with Protools, in which all the synth parts are recorded as MIDI tracks. I’ve been working along these lines for a couple of years, and feel I now have a fairly decent novice level of experience with it. I especially like the adaptability of editing compared to audio. Where I think I now need to learn more about is with the various machine controls.

As described above, I use the synth for almost everything, and all mixing is done in Protools, so there’s really nothing else in the way of gear other than the controller, my guitars and amp, and a pair of rather smallish monitors. My studio looks somewhat anemic compared to those of some of the gear junkies with pics in the “let’s see your room” thread.

I’m curious as to any consensus about the Roland JV1010. It seems to me to have some strong points and some weak. I was amazed when I first got it how many voices it has (thinking how would I ever need more than that?), but gradually came to realize that it’s not really all that vast. For one thing, some of the voices just aren’t very good, in particular individual string instruments. It seems a lot better on individual instruments with winds and various keyboards, and generally better with multi-instrument voices. Drums are usually acceptable for what I do, although I sometimes am wishing for a particular snare or tom sound that just isn’t there. Conversely, I rarely need all that Latin percussion stuff. Of course there’s a huge array of purely synthetic voices, but again for me is of more limited need. I’d be open to any general advice on the application of synths for the kind of recording I’m doing.

Another issue for me, that is a major limitation, is the complete inability to record any kind of vocals. My voice sounds very nasal, and I wouldn’t be able to stay on-key to save my life. Every once in awhile I try to record some singing, but always with the same result: “delete selected track”. Many of the original tunes I’m trying to make projects of have prominent vocal parts, so I have to arrange each of my versions to come up with some kind of alternative. I’ve used different synth voices as substitutes for the vocalist on various projects, ranging from individual instruments sounds (usually one of the winds), some string-ensemble sounds, and some of the “VOX” synthetic sounds. This approach works to some extent but creates a significant departure from the original tune, and I have found to many listeners results in the dreaded MUSAK character. Again, I’d appreciate any advice for improvement.

Also more by way of commentary here, I am forever dumbfounded by the MIDI program-numbering system! A more complex, arbitrary, and archaic arrangement I have never encountered (and I’m a scientist by profession and used to dealing with some complexity). I’m also struck by the phenomenon of the apparent dichotomy/rivalry between MIDI and audio recording. Seems to me that both sides have their merits, even though for my needs MIDI fits most of th bill.

OK, that’s plenty for now. Thanks for reading, and I hope to hear from you.
 
Welcome aboard Blackscot! That's a good narritive, and it sounds like you're having fun. I'm also pretty new to MIDI. I've been recording and working with audio for quite some time, but I always avoided MIDI as I was under the impression that it was all for Flock of Seagulls, and The Cure. I have recently come to the realization the MIDI is much more than cheesy synths, and actually really cool! I don't yet have enough knowledge to direct you at all, but wanted to say welcome anyway.

BTW: I use Pro Tools also, with the MBox2. Although I do most of my MIDI in Reason.
 
lucky13 said:
Welcome aboard Blackscot! .....I use Pro Tools also, with the MBox2. Although I do most of my MIDI in Reason.

Thanks lucky13.

My Mbox (1) came with some bundled software including Reason Adapted and Sampletank, which I tried out at first but seemed more limited than the Roland hard-synth and pretty quickly went back to. I noticed all the bundled stuff are "le" versions, which I wonder if are included just as a promo to get you to buy the full-blown. Maybe I didn't give them a fair shake tho....
 
Actually, as a computer scientist, the MIDI protocol makes sense to me, and is actually astounding in it's forward-looking design. Consider that it has endured for 25 years (an eon in computer technology).

That said, I've been using it since the 80s myself, so it holds little mystery for me. I suppose it would be a little intimidating at first to those who are not used to being exposed to computer protocols at all.
 
fraserhutch said:
Actually, as a computer scientist, the MIDI protocol makes sense to me.......I suppose it would be a little intimidating at first to those who are not used to being exposed to computer protocols at all.

I'm certainly no technophile, and am actually a geologist by profession working in the water-resources field. I see worlds of significant differences among what to most people looks like simply dirt, but likewise gnash my teeth at things like trying to get a formula copied down the column of a spreadsheet (actually, I'm doing it now....)

So, complex systems - yes, but definately of a different character. With MIDI I just wish there was an easier way to find and specify the voices I'm trying to use.
 
Are you using PT to sequence? If so (ot having used PT as a sequencer myself, shudder), there ought to be a way to specify a patch script/map that would allow you to pick the synth patch by bank/name.

Any PT guys care to chime in?
 
blackscot said:
With MIDI I just wish there was an easier way to find and specify the voices I'm trying to use.
Well, since MIDI is based on 8bit binary numbering, most of the values are going to be in the 0-127 range. Makes perfect sense for the computer, but it's not a nice round number like 0-100 that we humans like so much. To compound things more, some manufacturers have come up with their own way of numbering their patch/bank management system, in attempts to make it more "human friendly". Unfortunately, this can lead to issues when controlling one instrument from another if you're not careful.

Singing: The obvious answer would be "vocal lessons" :) Also, since you say your voice tends to be nasal, you may want to look at mics that de-emphasize higher frequencies specially around 4-5kHz... which probably will rule out most chinese mics :D

As to the MIDI vs. Audio rivalry... personally I don't see any. Each has it's own requirements and needs to be addressed differently. In the end though no matter how a track starts, it's going to end up as audio. Personally, when working with synths, I work the following way. I record the MIDI performance, tweak if necessary, then record the audio output. This way I can use another sound with it's own FX. I always keep the MIDI performances (just mute the channel after it's been recorded as audio), not only as backup, but also it's great to have the same part played by another sound, maybe doubling the original, or maybe coming in at a different point in the tune, to sound both familiar yet fresh at the same time.

Once in audio, I can then apply EQ, compression, or do some extreme editing, which would only be possible in audio.

As for drums, I would strongly recommend a software sampler or ROMpler as they will run circles around any hardware synth/workstation both current ones and the ones of the past. Battery 2, BFD, DFH, and others.
 
blackscot said:
Thanks lucky13.

My Mbox (1) came with some bundled software including Reason Adapted and Sampletank, which I tried out at first but seemed more limited than the Roland hard-synth and pretty quickly went back to. I noticed all the bundled stuff are "le" versions, which I wonder if are included just as a promo to get you to buy the full-blown. Maybe I didn't give them a fair shake tho....
They pretty much are dumbed down versions, bt the Reason Adapted has a lot of the instruments that are available on the full version. Although not even close to all of them. The good part is that with Reason Adapted, it's only $199 to upgrade to the full version, as opposed to paying $350 for it.
 
I share a similar time-frame in my introduction to midi and my knowledge of it. Learning about midi over the last few years has completely changed the way I think about music. Instead of the Roland, I chose the Yamaha Mu100 which has all kinds of bells and whistles, but as you said falls short in many ways.
Nowadays, I'm looking at the softsynth approach more and more as it easily integrates with my midi sequences and maps. As for Roland patch maps, I know Cakewalk/Sonar handles that realitvely easy in the form of instrument definition files and uploading sysex information. Try a demo!
As for singing, I've heard an awful lot of music where the vocalist wasn't actually singing as much as he was screaming or just plain hollering. You can do that, right? Heck, I remember a Michael Jackson album where Vincent Price literally spoke the entire lyrics!
 
ProTools MIDI sequencing is really a pain in the ass. I use pro tools but i have used Sonar before, and its really much easier. I use a Mac so sonar is not an option for me...

Just get a T808 or 909... hehe...
 
mikemorgan said:
.....I remember a Michael Jackson album where Vincent Price literally spoke the entire lyrics!

"Thriller"? Ever see the video? Best thing Jackson ever produced.

Thanks for the feedback all. Looks like some good leads for me to pursue. I'm aware of the record-MIDI-to-audio-then-tweak-sound technique. Haven't tried that yet but need to. Would seem to open-up a whole separate realm of control. If I'm still feeling limited maybe I'll think about one of the full-blown version softsynths.

As far as the singing, it really is hopeless (you need to hear it....no, actually not), and no amount of lessons and/or speaking/screaming is going to make enough of a difference. I need to just put down the mikie.
 
Back
Top