Amberwolf

Cloudwalkers https://amberwolf.bandcamp.com/track/cloudwalkers This is one of the very few that was made for a specific "story": I made it to be about a zeppelin voyage and battle in the clouds of another planet's sky, from a story "Last Cruise of the Zeppelin Tempest" by JP Boyd. So, fly thru some clouds and enjoy. All the guitar on this one was played in by me (I think on the Ibanez 6string bass), edited for mistakes as much as possible, and timing where I could. But that was five and a half years ago and I was still learning a lot about the many audiosnap functions (and related ones), so it could be improved. (also could be improved by having a real guitar player). EDIT: Looked up some notes, and it's not the Ibanez, its' a crappy First Act guitar. Pasted the notes in here in case they're useful or interesting. I have four guitar tracks. The first is the one I laid the basics of teh song down in, where i just played until I had something useful, and cut several minutes of the beginning out entirely, reducing it to the 9 or so it is now. Recorded in dry via the Avid FastTrackProUSB. In SONAR's effects bin for that track it has Wusikstation's FX, set to preset 16, with some of the levels of controls tweaked (don't remember what I adjusted, would have to compare the defaults to what I have to figure that out; I just tweaked till it sounded like I wanted). Next is SONAR's SonitusFX Delay, which is basically jsut a stereo delay, set to sync with the host with a left delay of 1 beat and a right delay of 3/4, which at the 99bpm of the project gives me 600ms/450ms. There's a crossfeed of about 10% between the two, and a feedback of about 13%, and a mix of 67% wet. THen there's SONAR's SonitusFX Delay, set to the Big Blue preset (kind of a big hall) tweaked up to be fairly diffuse. This track's main audio clip is also cloned to the Bass track, which has the SonitusFX Delay in it's bin with basically the same settings as the above, except only about 10% feedback and 6% crossfeed. The bass has a sub-bus it goes to that has the Sonitus Multiband EQ in it, which is setup for Vintage Compression and the EQ maxing out the bass up to about 80hz, then the next band up to about 300hz a few dB above normal, then the rest of teh bands way down below normal or zeroed. Effectively this could just be put in the bass track itself, as this song has no ohter stuff routed thru it, but it's part of my template so I just use it this way. The "fat" lead guitar track uses the same effect types as the first track, in it's own bin, except it has the VST BJ Overdrive between Wusik and Delay. Deptha and volume are maxed, with drive at about 45. The reverb settings are almost the same, but the early reflections are set much lower. The overlay track (the high pitched bits here and there) has the WusikStation FX with teh same preset as the tohers, then teh SonitusFX Modulator set to "Creeping Feedback" which is a string phaser, I tweaked it but I think most of the settings are the same. Then the same delay, but with feedback at 25% and mix at 30%. I didn't adjust the tone or volume controls on the guitar for any of them, both are at max clockwise, so full high tones and full volume. All of those except bass track feed to an Audio sub buss, whihc has the SonitusFX Multiband EQ set to th Vintage Master preset, untouched. All the busses go to a Main Buss that has another SFX MEQ, which I think was originally the VM preset, but I've tweaked it here and there enough that I don't recall where it started--but it is still very similar to that preset. It's pretty flat EQ, average compression, mostly just there to ensure the final mix doesn't distort but still "fills" the speakers. Without it, the audio is more dynamic, but must be turned down significantly to keep it from peaking out in a number of places in almost any project. (note from today five years later; not sure if it is, but it's what I did then) The best way to use a final EQ/comp like this is to automate it so it's bypassed almost all the time, and then just adjust it's mix ratio (or knees) so durign the too-loud parts it is compressing sufficiently to prevent clipping, but that most of the time it leaves everything dynamic. It's one of those "someday I'd like to go back and...." songs, that's about good enough as is, but could certainly be better in quite a few ways. Tempo changes would probably help it a lot. 😳 Haven't made new track art so it still has the old logo art

In These Amber Days https://amberwolf.bandcamp.com/track/in-these-amber-days Another experiment (from last year) that sounds like a medley of different things, I suppose (but isn't). Most of it is synths, but there are some guitar parts played on an Ibanez 6string bass (and then edited for timing, pitch, etc since I can't actually play, to create the parts you hear). Haven't made new track art for it, so it still uses the old generic logo

@jack c.@Wookiee (and anyone else following the thread that hasn't posted) It's been updated with some general edits and additional backing sounds. Same link: https://amberwolf.bandcamp.com/track/ookami-no-kari-no-yume-wolfs-dream-of-the-hunt There will be more updates as time goes on and I figure out how to do various things (every song is a learning and discovery process). Still working on the wind-instrument parts to augment the synflute-like sections, they're difficult to work with to get the sound i want, so none of those are in there yet.

Yup, that fixed it! Thank you very much.

In CbB (wahtever the latest is) it doesn't poof, AFAICT, (but the identical actions / tests in the old SONAR on the same test laptop does poof) so it's something about how the older SONAR handled buffering. I'll probably just have to avoid using the loop/AS in any projects I'm using MM in.

Thank you very much for the attempt...but the new version still POOFs reliably under the stated conditions on both my test system and my main one. Attached is an example project with the MIDI all within the loop, where it doesn't poof. But drag the clip past the loop end even a note's worth, and POOF once it hits the spot close enough to the end of the loop. I'm going to try to install CbB on the test laptop and see if it also happens there. viramor test 9c.cwp

The point I was making (and I think what the OP is) isn't about the automation system itself (which does have a channel picker in the envelope creation dialog)...it's that if you are attempting to assign (midi learn, etc) a control in a synth, there isn't a way to pick which channel that control is on, so you can't use the same CC (regardless of source) from different channels to do two different things. So, if you have for example two faders (say, on two different hardware modules) that are both locked to the same CC but you can set each module to be on different channels, then you cannot use them to control two different things in the same synth. Like if you wanted to use a built-in non-assignable volume fader (CC7?) on keyboard 1 to control resonance, and the built in non-assignable volume fader (same CC#) on keyboard 2 to control delay time, etc. It should be possible to remap a CC with an MFX plugin (if one exists; I don't seem to have one that can) so if you have this situation, you can put one of the faders on a different CC, so that it doesn't matter about the channel. EDIT: A quick google finds this post http://forum.cakewalk.com/Midi-CC-Conversionrerouting-in-real-time-m3611481.aspx pointing to this MFX by Tencrazy that should do it. (I thought I had all the TC MFX, but apparently not that one; I do now ) http://www.tencrazy.com/gadgets/mfx/TcMfxCcMap64-14.zip http://www.tencrazy.com/gadgets/mfx/TcMfxCcMap14.zip EDIT2: I tried to test that one, but for some reason even though it correctly registers, SONAR doesn't see it. Probably something about my setup since I seem to get other wierd issues; I'll keep poking at it. I also found that MFXScript has a USER script called cc2cc, which probably does this job, but I have not tested it, either.

You can have data recorded on a track in clips that sends different channels different data and a VST synth or external hardware that is capable of using that should respond correctly. But AFAIK you can't assign different controls on a VST (etc) to be automated by different channels of the same CC/etc. That's what the OP wants to do, AFAICT. I'd love to be wrong on that one. (I don't need the feature at present, but I can imagine there are people that do).

Minor MidiModulator UI bug: the numbers at the top of the Freq slider is only two digits wide, so anytime you go to a higher value it then show only the leftmost two digits, which is confusing. Since the result / output is still correct, this is just cosmetic and not very important. once the user realizes what is happening. The other sliders don't use more than two digits so they don't reveal this issue.

Behavioral change request: In MidiModulator UI, changing numbers in the CC boxes is more difficult than necessary. Except in one case, the user must first click in the box with the number to be changed, then select the text "OFF" with the mouse (can't be done with the keyboard and arrows, because the cursor is kept forced against the right edge of the box past "OFF"), then type the number you want. If you want a different number, you have to repeat this procedure, because as soon as you backspace or delete key the existing number out, you can't just type a new number as it now changes to "OFF". Additionally, if you have already selected "OFF" text, and accidentally type a non-numeric character, it unselects the text and you have to do it again. If you click in a box with a number, and don't backspace, you can type a number, but it is just inserted with the existing one, and the cursor is forced to the right side, past all of them, so you can't just hit "delete" (or arrow over to the one you want to go away and backspace), you have to backspace thru all of them and then your'e back to "OFF" and have to do the selection/etc. Better behavior would be to just replace the number or "OFF" in the box with whatever the user types once the box is clicked on to do this. The one case that's different that is already the desired behavior is if you click on the first box, then tab to the next one with the keyboard, and it will preselect the "OFF" text and replace it with whatever number you type.

In my ancient (couple decade old?) SONAR version, I never found the option to do this either--Any CC (etc) from every channel on every port is received the same; can't be assigned differently within any specific destination. (I didn't explore it much, as I have never had a good control surface, just one of those really cheap and crappy Worlde things, whose controls are not reliable or easy to use). Hopefully someone else has different news on later versions.

Testing the new plugs, I found that if loop/autoshuttle is engaged in the transport, CSMidiModulator will cause SONAR to just POOF, no crashdump, as soon as the sometime before loop/autoshuttle hits the loop end point. I suspect its at the point where the buffers for the plugin are hitting that point, something about how SONAR is feeding them during looping (I've had issues of various kinds with stuck notes or missed notes with various synths using looping, over the years, but not poofs) but I don't know how long it's buffers are. It happens so fast I can't be certain of the time, but I think it is at 3 measures into the 4 measure loop, which is 7:06 seconds out of 9:18 seconds. My MIDI buffers are set to 250ms. I haven't yet tested the results of changing buffer size, other than that it still happens whether I increase that 10x or decrease it to 1/10. Doesnt' matter if the MIDI is groove clips or just plain clips, or if it is one or multiple clips. It doesnt' happen if the MIDI does not extend beyond the loop end point. Only poofs if there is MIDI beyond the loop end. Happens even if all of the MIDI is outside the loop; there just has to be some in the track within a measure of the loop end time (past it). If it is further than that, it doesn't poof, and will sit there looping on nothing fine. (further testing shows it stops happening if the midi is within about an eighth before the end of that measure, but sooner than that and poof). The further the MIDI is beyond the loop end, the closer the nowtime will reach to the loop end before poof. It happens even from a blank starting template, creating any MIDI note(s) and placing them within the above constraints, or using existing MIDI clips imported in, etc, as long as loop/autoshuttle is engaged, within above constraints. Doesnt' matter where the loop selection is in the timeline, as long as the MIDI is within the above constraints to the loop selection. It doesn't happen if it just hits the end of the MIDI in a track and the transport stops on it's own, with no loop/autshuttle engaged. It doesn't happen if there is no MIDI in the track (even if there's MIDI in other tracks). It doesnt' happen with any of the other plugins, just the modulator. If you need further info, let me know and I'll do the tests.

T T # Nex Ool (read as tee tee hatch nex ool ) https://soundclick.com/share.cfm?id=14844484 was previously started a couple days ago here New MIDI FX Available That core track, anchor to the rest, used an arp pattern I'd made for something else, stuck into a midi track and looped just to play with the CSHumanize settings to see what they did, feeding a random patch in Z3TA2+ with it's own arp turned off. When I got something interesting, I dragged it out a bunch of measures then recorded it "live" with a MIDI loopback into another track (partly for the test, partly to see what it would do, etc). After muting the original track, I turned the arp in the synth back on (a simple up/down 2-octave four note pattern) and the track got significantly more interesting, so I started editing it as if I was going to make a song...and so I began actually making one. Percussion is two different kits (SI drums with the aggressive kit and SessionDrummer3 with my modified MisfitKit I like to use), both playing parts of the same clips (taken from the Aggressive Drums pattern in the SI drums selections, split to separate tracks / clips per drum sound, etc). Eventually I'll build my own percussion tracks based around this that specifically fits each bit of the song, but it's just this for now. I stuck a separate CSHumanize instance on each of the hats/cymbals/tamb parts (the tamb is copied and modified from the hats clips), mostly modifying the velocity but with a teeny bit of the note starts; the tamb only modifies the velocity, I think. ) Uses Perfectspace and one of the canyons modified as an audience/performance space (don't recall waht I did, was quite a while back; also used this for parts of Neotenous Chordata's bass fx track) to liven up the drums, along with a modified FlimsyKit in PX64Percussion. The SI kit is bussed right to the "percbass" bus (where I compress both bass and perc so they play off each other and cause less pumping of the rest of the tracks) the other goes into the drum bus first then to percbass. Each output bus of the SessionDrummer has it's own short fx chain, some pretty clean and some with reverbs, etc. The "sound effects" track has another perfectspace on it with a different large space and a multiband eq, which later on will be automated to narrow some parts down a lot to clarify the song itself there. Anyway, I'm still working on it, it's hardly even really started yet, but it's already listenable, and it's fairly short (only around 3 minutes). It will probably get longer as I build on sections, make breaks, etc. The best part for now starts close to halfway thru; the previous part is buildup (and is what is being worked on at present).

So far, it has. I'll letcha know if that changes. I'll test the new ones shortly. EDIT: several hours later, I didn't get that done yet; I was still playing with that test project, which now has it's own thread since it's going to be a real project track for sure now. The first set of 32bit ones (that crash on the main machine) also crash the same way on the spare laptop with a clean setup of SONAR 32 bit, patched up to 8.5.3. Presumably the same crashdump info, but I can post that if you can use it. Hardware and OS version are the same as the main laptop. Neither has the bloatware usually preinstalled on these, thankfully they came fairly cleanly setup, and on the test laptop I just did the default install for SONAR and the patches (other than making sure it's still in the PF folder rather than PF x86, mostly to make it easier for me to find things as I'm used to looking there, and to ensure compatiblity with the few ancient things I have that don't know about the x86 version of the folder path and don't work with it (which might be why I used the regular PF in the first place--it's been so long I don't remember). The DS_Test version that works on the main one also works on the spare laptop. I'll test the new ones on both, too. A thought: Are there any services that gdiplus.dll depend on? I have a bunch of services set to manual or outright disabled, to greatly speed up the computer and reduce glitches in recording and playback from them doing things I don't need to ever happen.... It doesn't seem to work for me, but I don't recall that feature in the ancient SONAR I have. I only have the freeze option (there aren't any rendering ones, other than exporting a selection to an externally saved audio file). If I drag the audio to a midi track, it just sits there as an audio clip and the track doesn't do anything with it. Would be a great function to have, though, so glad they added it somewhere along the line. I'll see if I can install CbB on the test laptop since it doesn't have anything to break yet, and then see if this can do what I'm after. But I suspect it won't without more extra work, because the synth sounds I am usually using don't always make exactly the same sound, volume, pitch, etc., (even with effects off) with different notes (number or length) being input, which is part of why I use them this way--they don't sound so mechanical. (they do make the same output for the same input data, making them repeatable, just not with different input data). So unless I use a patch that just makes like a sinewave at the same amplitude for whatever velocity input, regardless of any other note input or control info, any audio to midi conversion will give different velocity (and probably pitch) output than what I actually input, even without the internal arp running. I'll have to test to see. FWIW, There is an option in the audiosnap toolbar to create a midi frle from the audio, but it won't convert the pitches, just the timing and velocity. I wish CAL still worked reliably; I could probably figure out how to make or adapt a CAL to do this, but I was never good at it anyway. 😊 While it's a thought, and is worth a try anyway: For me, VVocal tends to crash projects, and doesn't seem to work on anything longer than a second or two (usually because of a crash, sometimes it just doesn't read the clips correctly or completely). It may be able to do this kind of conversion, but it would have to be done (and then checked to make sure it's correct) in lots of very short segments, probably hundreds of them; I'm not sure it would save me much work. Yeah, there are a lot of arp plugs, some of them with synths in them, etc., but... The old SONAR I've got doesn't support VST MIDI *output* stuff AFAICT. (it doesn't appear to even support feeding the onscreen controls from VST that have outputs for those back to the tracks to record automation from...if it's supposed to, then I must be doing it wrong--I wanted to use that function to save the live changes to Nerds.de's DelayLama VST some years back, and found I couldn't, though Nerds said DL supports doing it). If it does support it and I figure out how , then as long as the VST arp will output MIDI to a MIDI track that I can then record (or bounce, etc), to get me the final actual notes that a synth would play from "two arps in series" (functionally how what I am doing works), then that would accomplish my goal of getting a fully editable track full of notes. I'll poke around at it and see what I'm missing to try to get the VST MIDI plug(s) to work, on the spare test laptop setup.

Thanks! This specific piece is one of the ones I'm more proud of than the others; it turned out a lot better than it's original, linked below. https://soundclick.com/share.cfm?id=14566886 But that original abandoned idea did turn out to be usable to make this one from, which is why I save everything I play in, even if it feels like garbage. Sometimes it won't become a song on it's own, but I may borrow bits out of it to complement a song that doesn't have enough to complete it or is missing that one thing somewhere. The cover art, other than the lettering (done in MSpaint), was each a few hours of fighting with google's AI Test Kitchen to get it to do things that vaguely resemble what I was after. None of them are what I really wanted, but it would take me at least a month (if I didn't have a dayjob) to create even one of the ones I did get, doing it myself in GIMP or similar programs. AI doesn't "understand" anything, and the google stuff doesn't do iterative creation where you can tell it to make changes to an existing result and re-generate it. Instead it completely generates a new thing even if you just change one tiny word (or often, none at all), even if you lock the "seed" number in it's settings. (there is an edit function, but all you can do with that is give it a round spot to mark where you want change to happen, then tell it the change, then it really crappily shoves something else that's not what you told it in there, in a much worse editing style than even I had when I first started working with paint programs back in the 90s. But if you try long enough, eventually it can give you something close enough to what you want to be able to use, if you are not picky about details (like having five or six limbs, or three eyes, or instruments with 3 strings or keyboards pointing the wrong way to be able to be played, or cellos being played with drumsticks and vice-versa. 😆 The stuff I used to do myself in paint programs was like this, but I just don't have time for it anymore. (the moon and columns I created in TrueSpace2, and the nebula and exploding star were separate images from Hubblesite, IIRC; all blended together in PaintShopPro4).