KVR Audio

nikp2000 · Post by **nikp2000** » Tue Sep 12, 2006 2:48 pm

_

admiralquality wrote:If that IS the case, then you need to thin it FROM the GR-20 itself. Because once it's in the MIDI Cable, the delay is already there.

If you're inserting something to do "thinning" downstream of the GR-20 itself, whether hardware or software, it's not going to help one bit... as any messages are already held up by the time they get to it.

The synthesizer software in your computer is NOT having delays caused by any MIDI traffic, as even a saturated MIDI cable represents only the tiniest trickle of data to your computer, absolutely nothing compared to even one channel of digital audio.

I guess I'm really talking about glitching and false notes here, not delay. For me the main causes seem to be hitting strings with my picking hand and maybe resonant tones from other strings. All of this can be controlled with good techniques, sometimes at the expense of percussive guitar techniques. I've gotten good response delay-wise, but keeping unwanted notes and glitching under control is another matter that can be helped by clean playing.

My point on the amount on controller info coming to the vsti causing glitching comes from this train of thought:

most vsti expect a keyboard controller sending the midi info, which will be a much simpler stream, ie a note on message followed by velocity and control ccs, while the gr20 sends a combo of noteons and pitchbend. If the vsti expects keyboard-like midi data glitching may occur.

Now I'm not a programmer and I could be all washed up on this, but it seems logical. Anyway, a midi guitar designed vsti is way past due.

FYI, I started with midi on a commodore 64, then moved to an atari st before moving to pc. PCs are wonderful for audio, but midi functionality is way behind old midi only programs like Realtime.

Nick

_

AdmiralQuality · Post by **AdmiralQuality** » Tue Sep 12, 2006 3:55 pm

Again Nick, any amount of MIDI won't be a challenge for any VSTi (unless it's horribly programmed, which I really doubt.) Can you send me a sample .MID so I can see? Let me know what your guitar's pit

AdmiralQuality · Post by **AdmiralQuality** » Tue Sep 12, 2006 4:36 pm

Hey everyone, I was just reading some more of the GI-20 manual (download it from Rolandus.com!) and it HAS pitch bend and CC thinning functions (page 31, 32) though, as it has a USB connector, it hardly needs it! Still, nice to know it's there. Would come in handy controlling hardware synths.

ALSO... it supports different pitch bend ranges PER STRING! (page 24) Or they can be set all the same too. And it also sends MIDI messages to the synth AUTOMATICALLY SETTING THE BEND RANGE! (page 23) Only for synths that support that, but still, very nice feature!!!

GI-20 manual wrote: The messages used here are MIDI RPN (Registered Parameter
Number) "Pitch Bend Sensitivity" messages (Control Change
messages No. 100, No. 101, No. 6, and No. 38).

I want one!!! (And a Godin xtSA to play into it.

)

selfconstruct · Post by **selfconstruct** » Tue Sep 12, 2006 4:48 pm

There is also the "Chromatic Type" pitch bend settings which are quite useful for glisses and ligados. These work well with keyboard sounds (and can also be applied individually).

AdmiralQuality · Post by **AdmiralQuality** » Tue Sep 12, 2006 5:01 pm

Yup!

Do you have a GI-20 selfconstruct? If so, would you enter my contest? http://www.kvraudio.com/forum/viewtopic.php?t=147535

selfconstruct · Post by **selfconstruct** » Tue Sep 12, 2006 5:23 pm

AdmiralQuality wrote:Yup!

Do you have a GI-20 selfconstruct? If so, would you enter my contest? http://www.kvraudio.com/forum/viewtopic.php?t=147535

I do have one, and I will enter your competition soon. Actually I'm planning on buying a VG-88 this weekend so I'd better record something before then (I think I'll be very busy tweaking patches next week).

Sascha Franck · Post by **Sascha Franck** » Wed Sep 13, 2006 12:51 am

AdmiralQuality wrote:Hey everyone, I was just reading some more of the GI-20 manual (download it from Rolandus.com!) and it HAS pitch bend and CC thinning functions (page 31, 32) though, as it has a USB connector, it hardly needs it! Still, nice to know it's there. Would come in handy controlling hardware synths.

See, the data thinning isn't there to prevent you from bus/data overload. It even hasn't been there for such purposes in the past (ok, well, on my Atari the 1 MB of memory filled up quite quickly whenever I've been noodling around, so data thinning was a welcomed effect as well). But it's usually there to prevent you from unwanted notes (glitches and the likes) and pitchbends. If you wanted your host to do that job, you'd most likely have to use some offline processing. Ok, some hosts offer realtime MIDI input transformers (Cubase and Logic do), but they're tough to be optimized for a typical MIDI guitar signal (which would contain quite some PB and unwanted note data), whereas the algoryhtm to be found in most converters is quite optimized (sometimes even on an adjustable base) for guitar playing.
Even on my GI-10 (quite a mediocre device), when I activate data thinning, switch to no PB messages and probably also set the global sensivity a bit lower, the performance (well, what you hear) is getting noticeably better. Of course data thinning isn't appropriate for all playing situations. Expressive solo stuff requires it to be off (and PB to be on, of course). But for, say, playing straight ahead chord tracks, data thinning plus PB/off is doing quite a nice job.
Really, it's got nothing to do with bus speeds, just with filtering out unwanted data as early in your chain as possible.

AdmiralQuality · Post by **AdmiralQuality** » Wed Sep 13, 2006 12:53 am

Disagree. No data are unwanted.

Also, if you're turning off PB, what data is there to THIN? You sending some other controllers?

You're just misunderstanding the feature (again!) It's there to keep congestion on the MIDI CABLE down (or to save memory space on old limited hardware sequencers). NOT in your computer/host which has enough power and memory bandwidth to handle 10,000 saturated MIDI streams.

AdmiralQuality · Post by **AdmiralQuality** » Wed Sep 13, 2006 1:02 am

I just read the GI-10 manual, and it's a BEND data thin feature. So I was right, if you're turning Pitch Bend off anyway, there's NOTHING TO THIN! (Page 14 of your manual.)

You were confused again. Live and learn.

Sascha Franck · Post by **Sascha Franck** » Wed Sep 13, 2006 1:14 am

Even if you are right that it's only bend data thinning, you obviously never played a MIDI guitar. Data thinning these days is NOT there to allow for proper operation with whatever ancient equipment. It's there to offer better performance - when it comes to bends, I simply don't want or need PB data on a piano track. And *each* note you play on a MIDI guitar *will* send PB information, due to the nature of the instrument. So you better filter them out before they even reach your sequencer.
And yes, I was confusing it a tad with global sensivity (which effectively results in some data thinning as well), but that shouldn't matter.
Data thinning is there to deliver some cleaner performance, and that's all there is about it. It's got *nothing* to do with the amount of data created.

AdmiralQuality · Post by **AdmiralQuality** » Wed Sep 13, 2006 1:19 am

Sascha Franck wrote:Even if you are right that it's only bend data thinning, you obviously never played a MIDI guitar. Data thinning these days is NOT there to allow for proper operation with whatever ancient equipment. It's there to offer better performance - when it comes to bends, I simply don't want or need PB data on a piano track. And *each* note you play on a MIDI guitar *will* send PB information, due to the nature of the instrument. So you better filter them out before they even reach your sequencer.
And yes, I was confusing it a tad with global sensivity (which effectively results in some data thinning as well), but that shouldn't matter.
Data thinning is there to deliver some cleaner performance, and that's all there is about it. It's got *nothing* to do with the amount of data created.

Stop trying to contradict me. You're clearly not qualified and you keep having to change your statement in (failed) attempts to not look like a dumbass.

A MIDI cable is no faster now than it was in 1983. So it doesn't matter how fast our computers are, if you're hooking up with a MIDI cable, then there's a good reason to consider data thinning.

The device wouldn't GENERATE a pitch bend message if it didn't want to. If your tracking is inaccurate, then you're in fact making it WORSE by thinning, because it'll be that much longer before the next, hopefully corrected, pitchbend message comes along.

Why filter them out before they reach the sequencer just because I'm playing piano? MY piano patches don't respond to pitch bend. But if I ever change my mind and want it, it would be there. But not if i throw it away before I can even record it.

And AGAIN, if I didn't want PB, I'd turn it OFF, NOT THIN IT.

And isn't it ironic that I don't even HAVE to play a MIDI-guitar, to know more about it than you? Funny how just because you own something, doesn't mean you know anything about it.

But you don't know anything because think poorly. Stick to guitar.

Sascha Franck · Post by **Sascha Franck** » Wed Sep 13, 2006 1:32 am

f**k you. I tried to get back to some reasonable discussion, but you're just behaving like the last asshole. Which you unarguably are.
I'm replying nonetheless. Thinning out PB data makes a LOT of sense when you deal with it in an editor. You're just too much of an idiot to see that.
Pitchbend thinning further results in less unwanted jumps of the actually recorded PB values. Again, you're too much of an idiot to even think about such things.
Had you *ever* played a MIDI guitar, you'd *know* about those unwanted PB glitches, and you'd also know about the benefits of PB data thinning - but all your wisdom is coming from some manuals you downloaded. Great. Continue with it, asshole!

kevink · Post by **kevink** » Wed Sep 13, 2006 1:33 am

AdmiralQuality wrote: Stop trying to contradict me. You're clearly not qualified and you keep having to change your statement in (failed) attempts to not look like a dumbass.
...
But you don't know anything because think poorly. Stick to guitar.

Are you on crack or something?

kevink · Post by **kevink** » Wed Sep 13, 2006 1:35 am

Sascha Franck wrote:f**k you. I tried to get back to some reasonable discussion, but you're just behaving like the last asshole. Which you unarguably are.
I'm replying nonetheless.

Error. See that little "mute" button in his author box? Just click it. He's not worth your time.

AdmiralQuality · Post by **AdmiralQuality** » Wed Sep 13, 2006 1:41 am

The asshole is the guy who feels he has to argue with every technical point I explain here. And please note how I haven't commented on any of the purely opinion based stuff he's said, just on the FACTS he's gotten WRONG. Particularly the ones attempting to refute some very good background information that I've chosen to take time to share with the members here.

Yes Sascha, please, mute me... then I won't have you contradicting me anymore about stuff you've repeatedly proven yourself unqualified to comment on. And you should try READING those manuals instead of putting down those who do, maybe you'll actually LEARN something.

Dumbass.

VSTI world hate MIDI GUITAR?