syncing an external keyboard with Sonar

Thanks very much for taking the time to respond. Your response has been helpful. I may respond to your points out of order, but I'll try to respond to all of them.
 
I'm well aware of latency. Fortunately my sound card, an M-Audio Delta 66, can achieve very low latency levels. I think that right now it's somewhere under 10ms. If I use the 1ms = 1 ft rule, this means it's the same as listening to my amplified guitar from less than 10 feet away. Good enough for me.
 
I haven't hooked the synth's audio outs up to the computer yet. Fortunately, my Delta card has a break-out box called "Omni IO" which is, essentially, a compact four channel mixer. So I'll be able to run my synth in through it.
 
I should probably back up and give you a clearer picture of my setup. When I built this machine, I built it as my DAW. But that was several years ago. We moved to a new home recently and I'm still dealing with boxes all over the place. I have not reassembled my DAW yet. So far, all I have going is the computer and this one keyboard. I have a couple of mixers, a few external effects processors, another keyboard, a Roland GR-33 guitar synth, and a Roland JV1010 synth, all of which I need to hook up to this machine. With the exception of the mixers, it's all still in boxes and I don't know which boxes all the stuff is in. So later on down the road, things are gonna get a whole lot more complicated than they are now. But still at the heart of the beast will be my trusty old Delta 66 card.
 
The Delta card usually defaults to 256 buffers. I often run it at 1024 if I'm getting some static in my output. But that doesn't seem to be a problem yet. I prefer running it at 24/96. 96k I've found, substantially reduces latency from slower rates, like 44.1k, for instance. In that little performance window, CPU tasking is pretty low. This is a fairly quick machine even if it's getting a little old now. I'm running Win7 64-bit, a 3.4 GHz processor, quad core, running 8 gigs of RAM, plus I have a huge amount of disk space -- 3 TB, little of which is occupied with programs and other data so far. I dunno when (or if) I'll be upgrading to Win10. So far I see no need for it. I'm perfectly content with Win7. It's the best OS M$ had ever produced, and that includes v8.x. But I dunno, maybe 10's better. I suppose if I were ever to get a laptop with a touch screen, I might want to run Win10. But I have an iPad for that.
 
The only way I can turn off this keyboard's speakers is to plug something into the headphones jack. One thing that would be illuminative would be to let the speakers be and play a piece with the kb running through my sound card's mixer -- see just how much latency is actually occurring. If under 10ms, it should still be noticeable but just barely. But the kb's earliness is approximating 125 ms, if I'm doing my mental math correctly, so I should definitely be able to see the difference. Tempo of the piece = 120 bpm or 2 beats per second, 1/4 beat early = 1/8 second= 125 ms.
 
I've brought this tune into Sonar as a MIDI type 1 file, and, as it sits, there are no effects at all associated with it yet. But I am using MS GS Wavetable Synth because that's the only one that comes up as a default. I don't like it much either. I like Cakewalk's TTS-1, but from what I understand it's just the GS Wavetable synth being run through a Cakewalk mixer, yes?. I have a few other software synths, but none of them supply the standard MIDI selection of instruments.  Well, anyway, I'll take a look at that, as well. IF I can get the Aria player that came with Band in a Box to work with Sonar, that'll give me quite a few sounds to choose from.
 
Hello Base 57, and thanks for your response. Will it make a difference if there is not yet any audio that Sonar is having to handle? Right now, it's all MIDI. Appreciate the tip, though. This could come in handy for when I am dealing with audio as well as MIDI.

OK I'm glad I read your second post as that explains a lot more about your system.
 
With what you are doing there should be no delay. I think I know what your problem is..your midi routing. 
 
Most of the set up for using external sound hardware is found in preferences. 
You pick your audio interfaces Midi ports MIDI/Devices/Outputs/ Delta midi
Now in the OUTPUT selection of the midi track you select that as your output.. or a VST synth. I know it can be done but I've never sent a midi track to both a VST and external. I always just duplicate the tracks for those situations. I really stopped using external sounds a long time ago because the quality is always sub par for what I use inside Sonar now. 
 
I think your issue is for some reason you also involving the MS wavetable... don't use that bl;oody thing, it's terrible. 
 
You are way better off using Sonars many VST synths. The TTS-1 will automatically load up when you OPEN any GM midi file , but to do this you would have to go back to MIDI/Devices/ output and make sure nothing is selected first.
That's a great trick I use all the time. The song will play the way it was meant to sound. Then I just replace TTS-1 sounds with better VST's.  You can always go back and check your interface midi output after the song loads up if for some reason you have a sound in your keyboard you just can't live without.  TTS-1 is a lot better than Wavetable. Wavetable is like 4bit.   
 
 

Could you set all the MIDI tracks to play through the keyboard?
I suppose it would be a problem if the patch you want to hear doesn't exist in the keyboard, though.
I record all my MIDI using my keyboard's sounds and later change them to soft synths. Or not.

I've been chipping away at the problem with my keyboard playing ahead. I tried something simple. I reduced the tempo of the song -- a lot. From 120 down to 80. What become immediately obvious is that it's playing ahead not by a tempo value, but by a time value. It's still playing ahead by a fractional amount, but now that I've slowed the tempo down, it's obvious that it isn't a quarter beat ahead. It's more like an eight of a second ahead. Sounds small, but that's 125ms and enough to drive one nuts.
 
So now I'm gonna go take a look at anything associated with time and not tempo.

As I said , you don't want to send midi info unless your using your outboard gear. SHUT ALL THAT OFF . 
Your probabley getting a midi loop back as the TTS_1 is playing, sending midi out and it's coming back again. 
 
Try this. 
Go to midi devices/output uncheck everything. 
OPEN a GM midi file. This should open a new project and insert the TTS-1 automaticly. 
 

Hey - here is another thought.  With ANY keyboard containing on-board (local) sounds, that I hook into my computer, via either USB or an old-school midi connector, I go into the Edit parameters for that keyboard, and turn OFF a midi parameter called Local.
 
What that Local parameter controls is whether or not pressing a key on that keyboard will have the keyboard react to that to trigger a note.  With Local Off, it is like separating the pressing of keys from actually generating a sound.
 
Now, on the surface, that would seem rather silly, but because you have the keyboard connected into your computer (and as a midi input device and a midi output device in Sonar Preferences), even though the keyboard won't generate a sound directly from the pressing of a key, it WILL still generate a sound, because that midi data is sent into Sonar, and back again through the midi connection to the keyboard, which will THEN cause that key's associated sound to play.
 
There are 2 reasons the above will help:
 
1) It will almost always result in sounds from other tracks and sounds from the keyboard lining up properly, with it all essentially being processed in the same manner from within Sonar.
 
2)  With Local set to 'Off', you prevent a second note playing the sound for a second time.  If you DIDN'T turn your keyboard's midi Local parameter to 'Off'. then what usually happens is that you would get TWO notes sounding every time you played a single key on that keyboard.  The first note would sound from the keyboard reacting to the pressing of a key, and the midi data going into Sonar and back into the keyboard again would then trigger a SECOND sound for that same key, with a slight latency making it sound like an echo. (for sounds that are mono, or for synths where there are limited voices, you could also either goof up the sound heard or actually end up with sounds dropping off, because of the keyboard processing each key pressed a second time - if Local was left 'On').
 
SO - the general guideline would be that for keyboards with on-board sounds, versus midi controllers that do not have their own sounds, you would always go into the Edit functions on that keyboard and set midi parameter 'Local' to 'Off'.
 
Using the above approach, you may well never need to mess with manual setting of the Automatic Delay Compensation that Sonar defaults to doing for you.
 
Here is a link that explains Auto Delay Compensation:
 
http://homerecording.com/bbs/general-discussions/newbies/what-exactly-automatic-delay-compensation-314029/
 
I hope the above helps you with your situation.
 
Bob Bone
 
 
 

Guys, the only time the Casio shows up in Sonar is when I have it on. And even then, if I haven't been using it, it will time out after a while and shut itself off.
 
Next time I start it, I will check to make sure that Local is OFF.
 

A bit of digging around and there are lots of posts going way back with the TTS-1 playing distorted and most it was people using higher sample rates. So you either have to revert to 44.1 or 48 or abandon the tts_1.
 
As an experiment try this. 
 
Go into your preferences and change the sample rate to 44.1 or 48. APPLY. 
Make sure your using ASIO mode and the Delta is your timing master etc. anything to do with your on board sound needs to be unchecked. 
Now make sure there is no devices selected in MIDI/ output. 
Now OPEN a midi file. Do not import. 
I have midi files set to use Sonar as the default program to open, do this or using the FILE menu choose open and browse for a midi file. Open the midi file and TTS-1 should load and play the file perfectly. Tell us what happens. 

mwmcbroom
"Horrendously out of phase"

 
Probably not that relevant but.. Have we established the soundcard type ?
 

I've been kinda busy the past couple of days so I'm just getting caught back up on all this. You guys have thrown so much stuff out there to think about and try, I haven't sorted through it all yet.
 
But I do have positive news to report. A couple of your comments led to it. First, I realized that it's been quite a while since I've checked on available updates, and turns out the version I was running wasn't close to current. So I installed the latest update last night. I also tried I guess it was Johnny's advice and reduced the rate down to 44.1. The system is still stuck at 24 bits, even though the front page of the song shows 44.1k @ 16 bits up toward the top of the page there. The section in Preferences where it shows my system set to 24 bits is grayed out and I haven't figured out how to make that active yet. So I'm running with a bit of a mismatch currently, but it doesn't seem to be affecting playback. I haven't tried any recording yet.  So anyway, I dunno if it was the update or reduction of the rate that straightened out the problem with TTS-1, but it's behaving normally now. Well, with at least one MIDI file that I brought in (and subsequently saving as a cwp file). With another, I dunno where I've gotten crosswise with TTS-1, but I can't get it to make a sound. Even though it seems that I have everything set the same between the two files. I guess I need to take a closer look. I've missed something somewhere.
 
Both Johnny and bvideo mention that TTS-1 isn't just a Cakewalk wrapper for the GS Wavetable synth. So I'm wondering why it sounds exactly the same? Or maybe I've misunderstood what y'all were getting at. I like Rapture a lot, but somehow it got corrupted on my machine, so I've d/l'd it again and reinstalled it, but I haven't gotten to try it out yet.
 
One question -- you guys keep referring to setting my kb to "local off." Is this a setting I can find somewhere in Sonar?  Because if it's one that I have to execute from my kb, then I've got a problem and I don't know what the workaround will be. There are instructions within the documentation for my kb on accessing MIDI setup and controls, but unfortunately my kb does not do what the documentation sez it's supposed to do. From the very git-go, where it tells me how to get into midi setup, a different message appears on my kb's display from that shown in my kb's docs (and it is the docs for my kb too), and none of the other keystrokes do what they're supposed to, either. So until and unless I can get some response from the Casio community on this, I'm stuck with whatever the default configuration is with that device.
 
 

Johnny, sorry if you misunderstood. I didn't write that you indicated that TTS-1 was a wrapper for MS Wavetable. I was writing that I thought it was. But what I did note is that it sounds the same as GS Wavetable. But I guess it's greater than 4 bit at least.
 
As for the clock rate, I prefer to run my Delta 66 at 96k, and I would like to do so in Sonar. Too bad TTS-1 can't handle it.  My recordings ultimately aren't "hobby" related because I will be producing CD-compatible Wave files. I realize CD files are 44.1k/16-bit, but I like to work at the higher rate, the convert downward after I've completed all edits to the music.
 
I'm gonna try opening a midi file again with the GS Wavetable deselected. First time I tried, TTS-1 didn't load itself. But that was with the older version of Sonar I was using. I'll give it another try momentarily.
 
 
 

I'm working on a fresh MIDI file, and I'd just like to report that, with GS Wavetable de-selected, TTS-1 came up fine. I like.