Re: OT RE: FC300 and VG-99

[Mech <mech@m3ch.net>]

Second forwarded message on the topic of the VG-99.  This one deals 
primarily with integrating the 99 into a laptop setup....

        --m.


>Date: Fri,  7 Sep 2007 07:53:46 -0700
>From: Mech <mech@m3ch.net>
>To: vg-99@yahoogroups.com
>Subject: Computer Experiments, Part 1...
>
>Sorry for the longish silence, but I got hit up with email troubles, 
>as well as
>being run over by RL in general.  And, of course, I've been trying to 
>spend as
>much free time as possible actually playing with the VG-99, as opposed to
>writing about it.  But, I happen to be on another overseas flight, so this
>looks like as good a time as any to update what I've found so far.
>
>Warning: Geekfest ahead!!!  ;)
>
>First, I did actually get my Vox back in the interim.  So, while 
>testing things
>out the VG-99 with a fretless 11-string is not exactly going to translate 
>well
>to the future experience of others, it is a little more applicable than 
>using
>my Chapman Stick (not to mention a bit more stable, FWIW).
>
>In general, I remain impressed with the sound quality of the 99.  I 
>continue to
>be a big booster, finding new ways of generating original sounds almost 
>daily.
>However, it's not without some pitfalls, and I'm going to hit upon a few
>warning notes with this email.  So, while the post overall is going 
>to be a bit
>of a downer, I need to reiterate that the VG-99 is still a very cool box,
>IMNSHO.  This will merely serve to list some aggravations, hopefully so 
>that
>others will not go in without being forewarned.
>
>To begin, as the Subject line suggests, I've spent the past couple weeks
>wringing out the bugs from integrating this new piece of gear into a
>laptop-based setup.  We'll go through a couple of minor 
>aggravations, then I'll
>spend the rest of this note addressing the biggest issue.
>
>First, if you haven't already discovered this yourself, READ THE 
>INSTRUCTIONS
>BEFORE INSTALLING THE DRIVERS.  Yeah, I'm on a MacBook, so I've been 
>spoiled by
>simply letting Core Audio take care of all that stuff in the background.  
>The
>MIDI configuration is a bit more complicated than the usual plug-&-play 
>we're
>used to, though.  So, read the instructions top-to-bottom, then follow 
>them
>carefully one step at a time.  If, like me, you're one of those 
>impatient types
>who normally just throws in the disk, boots the program, and runs... well,
>you're probably looking at trouble with your usual approach.
>
>That said, the MIDI actually works much better than I anticipated.  On 
>Stick,
>it's a mess, so I haven't yet tried to debug that (not even really 
>interested
>in using MIDI with the Stick, though, except as an academic excercise).
>However, with the Vox, it tracks!!!
>
>Now, in case you're not quite getting this, I'm using a *fretless* 
>guitar, with
>*doubled courses* detuned from each other for a chorusing effect.  This 
>thing
>should glitch from here to Sunday, and I wouldn't find fault with Roland 
>if it
>did.  However, the tracking is actually pretty darn good.  And I am NOT 
>using
>chromatic mode, which would normally just round the pitch to the next 
>whole
>step.  What I am really impressed with is that it tracks the pitchbends 
>very,
>very well.  In fact, when using AAS' String Studio, I can even play
>microtonally, as the VG is able to track the note AND the pitchbend 
>offset on
>the initial pick of the string.  Sweet!!!
>
>Next aggravation...
>
>This is a minor gotcha, but it will be familiar to those of you who 
>have worked
>a fair amount with USB interfaces in the past -- especially on laptops.  
>If
>you're going to be using the USB for audio, look at your system 
>configuration
>before plugging the VG into a USB port.  More than likely, your trackpad 
>or
>mouse will already be assigned to one of those serial busses.  Now, 
>there are a
>few masochists out there who actually enjoy spending hours on pointless
>troubleshooting.  If you are one of these people, then feel free to 
>plug the VG
>into the same bus as your trackpad.
>
>Myself, I discovered this by accident, since I've spent the past few years
>working with Firewire rather than USB.  I spent a damn long time trying to
>chase down these phantom pops, which I was erroneously attributing 
>to cache and
>latency settings.  Finally, I realized that the darn thing was only 
>popping and
>crackling whenever I touched the cursor onscreen.  Checked the settings 
>and,
>yep, I'd plugged into the same bus as the trackpad, so my finger 
>movements were
>feeding garbage into the audio outs on the VG.  I switched ports, and 
>problem
>solved.
>
>Finally, the biggie.  Go get a cuppa joe before tackling this one, as it 
>will
>take a while to outline fully....
>
>Originally, I had conceived of the VG-99 as a central "hub" in such a 
>setup,
>taking the place of a conventional external soundcard entirely.  The 
>simplistic
>elegance appealed to me; showing up to a gig with only an axe, laptop, 
>VG, and
>FC-300.  Power and a mains run to the PA run notwithstanding, such a setup
>would only require 3 cables in a perfect world: 13-pin to connect 
>the guitar to
>the VG, USB to loop audio/MIDI in & out of the laptop, and finally an 
>RRC2 for
>the FC-300.
>
>Such a simple setup is not to be, though, due to a seemingly minor 
>-- but fatal
>-- engineering decision (read: inadvertent blunder, IMNSHO) on Roland's 
>part.
>You see, when changing patches, the VG-99 temporarily mutes its output, 
>most
>likely to avoid any glitches or pops that might occur when changing 
>patches.
>Unfortunately, instead of merely muting the VG's synthesis engine, 
>it mutes the
>mains entirely for a half-second.  That also causes any extraneous audio 
>being
>processed through the laptop's USB connection to also be muted.
>
>So, if you are using your computer for backing tracks (drums, for 
>example), Live
>Looping, or even outboard effects (delay and reverb effects in 
>particular), you
>are going to have to deal with *everything* going silent for a half-second
>whenever you switch from one guitar patch to the next; longer if you need 
>to
>scroll though patches with the jog-wheel.  Agh!!!
>
>After discovering this sad fact, I've spent the past couple of weeks 
>working out
>a solution, but it does ruin the "simple elegance" of the 3-cable system I
>mentioned above.  The answer, as you might have guessed, is another 
>external
>soundcard for the laptop.  What I did was to try finding the cheapest and
>smallest decent-quality S/PDIF-capable interface available.  In my case, I
>settled on the Echo Audiofire2, which is basically a 2-in/2-out firewire
>interface with S/PDIF and separate headphone outs.  I take the 
>S/PDIF outs from
>the VG into the computer, then use the balanced outs from the Audiofire 
>into
>the mains.
>
>Why S/PDIF?  Well, I was also having a few latency issues with using the 
>USB
>feed.  We've never been able to officially determine if the USB 
>ports on the VG
>were v.1.1 or v.2.x compliant (aside: I'm currently leaning toward 
>version 1.1,
>FWIW).  In Ableton Live on a 2Ghz Core2Duo MacBook, my best latency 
>settings
>were ~10ms input and ~8ms output, resulting in a roundtrip latency 
>of around 18
>milliseconds.  On paper, at least, this is not good.
>
>Now, Live has latency compensation built into it, so it didn't 
>really feel like
>a full 18ms (which is actually a really horrible latency number by today's
>standards).  Now, I don't know about the setup for other DAWs -- they may 
>not
>necessarily be able to compensate for it.  And the VG just didn't feel as
>snappy and responsive as it should.  I don't use the Direct Monitor 
>function,
>FWIW, because I do Looping inside the computer and I need to hear the same
>timing that the computer hears.  As a result, it's much easier to be 
>affected
>by lousy latency numbers.
>
>So, by using S/PDIF, I was able to get around the USB performance 
>issue.  Using
>the digital interface on the Audiofire, I was able to whack off 
>around 5ms from
>that original number.  And using Live's compensation brings the "feel" 
>back
>into line with what it should be.  The only trade off is in the 
>clocking.  If I
>use the VG's clock from the S/PDIF, everything seemingly works great.  The
>exception is that if the S/PDIF cable ever gets pulled, the laptop audio 
>locks
>up entirely.  That's not likely to happen, but it's one more dragon 
>which could
>raise its ugly head and utterly fry you in an onstage situation.
>
>Alternatively, I could clock from the computer (essentially both VG and 
>laptop
>are unsynced with each other; the VG has no digital *input*, so it can't 
>sync
>to an outside source).  If doing that, I end up with occasional soft 
>ticks in
>the output -- about one every 5-10 seconds.  This is expected, since the
>waveform and sampling rate are not synced and this will cause a digital 
>glitch
>(tick) every few seconds.  However, the tick is really more noticeable 
>with
>careful listening, so it's possible that the best solution might be to 
>break
>the digital sync while playing live (where you may not even notice it) but
>enable the sync in the studio, where you don't wind up with an eternity of
>public humiliation if you accidentally kick the cable out of the back of 
>the
>VG-99.
>
>Oh, final digital note: the VG's engine is 24-bit, 44.1khz sampling 
>rate.  That
>won't change; no-how, no-way.  It's a bit of a minor aggravation, since I
>actually prefer operating my DAW at 48k -- too many years working with 
>DAT, so
>its just a personal preference -- or even 88.2khz, if my CPU can take it.
>Unfortunately, I've got to reconfigure my setup to accommodate the VG-99's
>digital quirks again.
>
>I've got a few non-computer-related VG-99 findings/warnings too. 
>However, since
>I've probably put at least half the list members to sleep with digital and
>laptop geeking, I'll list those in a separate post.
>
>       --m.
>
>
>--
>_____
>"I want to keep you alive so there is always the possibility of
>murder... later"
>
>----------------------------------------------------------------
>This message was sent using IMP, the Internet Messaging Program.