Robert Bone

Update - I still have to go through the examples I just found with a simple search in Google, but it looks like it is explained in numerous posts how to play it live. (why it never occurred to me to look on the web - though in my defense, I had been stumped for almost 40 years, so had quit looking by the time such explanations had been put out on the web). I wanted to update this thread right away, so I wouldn't have someone putting time and effort into answering my questions, when it appears the answers are indeed out on the web. I am about to venture off to look at them, and will update this thread with what seems to be the winning formula. Bob Bone

In interviews, early on, I remember them joyfully explaining the separate pitch bends in the studio, for each note, and I have remained stumped ever since. I am hoping somebody has solved this riddle. Bob Bone

Howdy - I have always wanted to play out, the Toto song Rosanna, but was always stumped by one aspect of the instrumental middle section of the song, and how to replicate it live - though THEY seem to be able to do it. In the closing section of the keyboard lead in the middle of the song, there are a series of pitch bends applied to chords - G major chords, in different inversions, starting with the root position of the chord, and bending each of three notes down to the corresponding lower pitch of the next lowest inversion, before then playing the next higher inversion and repeating the downward pitch bend back down to the prior inversion, etc.... SO - there never used to be a way to accomplish the above, outside of tracking each note separately in the studio, as each pitch bend had to move a different interval downward. For example, start with a simple G major chord, with intervals 1-3-5 and notes G-B-D. That then bends all 3 notes down to different pitch bend intervals - going from 1-3-5 (G-B-D) down to 6-1-3 (D-G-B), which involves the G bending down a 4th, the B bending down a 3rd, and the D bending down a 3rd. The next set of bends starts with the next highest G inversion (1st inversion) up from initial root form, with intervals 3-5-1 (B-D-G), and bending down to the root chord again 1-3-5 (G-B-D) so the pitch bends of the notes are the B down a 3rd, the D down a 4rd, and the G down a 4th. SO, each chord applies a combination of different pitch bend intervals for the 3 notes in each inversion of the G major chord, and was impossible to play live, on a single keyboard, at the time it was released. AND YET - if you watch any of the live performances of late, showing the keyboard solo being played, they are somehow able to accomplish those multi-intervalic pitch bends, and I would like to understand how I can replicate doing that for live performance. Apologies for the lengthy explanation of the issue - it has stumped me for years - does anybody have some sort of way I can accomplish doing this live? Is it some sort of polyphonic portamento? Bob Bone

Well, Kontakt is the king daddy in the market, with EastWest Play trailing a bit, but also good, but the EastWest sample libraries can get pretty pricey. The above being said, I also have TX16WX, and it is really a solid sample engine, as well, and you can't beat free. I have not tried their commercial Pro version, but I can vouch for them putting together a quality product, even with their free version. I do not know how extensive the libraries are for TX*, though, but why not give their free version a whirl, and see how you like it? That's why it is there, and lots and lots of folks use one or the other flavors of it, quite happily, from what I have seen. Bob Bone

From the doc: " "Mixing Latency (MME and WDM drivers only) Buffers in Playback Queue. This value determines the buffer characteristics for transfers to and from the audio drivers. Lowering this value improves audio Latency, though making it too low makes your system more susceptible to stuttering or dropouts. Buffer Size Slider. This control lets you set mixing latency manually, overriding the value set by the Wave Device Profiler. Lower numbers increase the risk of audio problems. WDM sound card drivers offer lower latency than the older MME type." So, try to adjust your ASIO Buffer Size - I generally use 128 samples when recording, and 1024 or 2048 when mixing. Curiously, what specific effects might your project have loaded into it? Some plugins are meant for use during mixing/mastering, and require a large ASIO Buffer Size to work properly. You can test that,m by temporarily bypassing effects processing, by hitting the letter 'E' on your computer keyboard, and then starting playback. If the issues instantly go away with the effects bypassed, then one or more effects are likely requiring a larger ASIO Buffer Size, so either turn them off during recording, and back on after shifting the buffer size to either 10243 or 2048 during mixing, or swap them out for different effects until you finish recording, and have moved on to mixing, when you can swap them back into the project, again once you also enlarge the ASIO Buffer Size for mixing, to a value large enough for those effects to do their thing. Convolution Reverb effects, and those that use Linear Phase processing, are usual suspects for this kind of thing. Hitting 'E' again, by the way, will turn the effects processing back on. Bob Bone

Do the crackles go away when you temporarily bypass all loaded effects, by hitting the 'E' key on your computer keyboard, and then hitting play? (you can turn effects back on, by simly hitting the 'E' again after the test). What Driver Mode is your project using? ASIO? etc., What Sample Rate is your project using, and what is your ASIO Buffer Size set to, if using ASIO Driver Mode? What kind of audio interface are yu using? (Focusrite, Steinberg, etc....) Bob Bone

What sample rate are you running with? With a Driver Mode of ASIO, I would think a Sample Rate of 48 K and an ASIO Buffer Size of 128 samples, should give you decent performance. Bob Bone

I have generally used AMD CPU's for most of my desktop PC's, for the last 10+ years, and love them. I have a RYZEN 1950X Threadripper, in my main desktop, and it is a beast of a machine. AMD CPU's do a fine job with streaming audio apps, such as Cakewalk by Bandlab - you get more bang for your buck with an AMD CPU, versus using an Intel CPU, though the Intel chips run many types of tasks faster - generally speaking, the AMD chips have better performance at the equivalent price point. I should point out that when I built my monster desktop, the AMD motherboards had zero support for the then brand new Thunderbolt 3 protocol. My laptop has an Intel CPU and Intel-based motherboard (of course), and it also has the Thunderbolt 3 support on it. I believe some of the newest AMD-based motherboards have support for Thunderbolt 3, but I do not know which models of motherboard have that support. SO, if you seek Thunderbolt 3 support - just make sure whatever you buy supports it. Have fun with whatever you build. Bob Bone

Do you have the Input Echo On button turned on, for the vocal track you are trying to hear while you record? It is just to the right of the Record button on each track, in the Track Pane. Bob Bone

Well - according to the help documantation: "WDM/MME/WASAPI output buffer not available for delivery to audio device", and the suggested action to resolve is: "Increase the Buffers in Playback Queue value in Edit > Preferences > Audio - Driver Settings". So, you are using a Driver Mode other than ASIO? Have you tried going in to Edit > Preferences, and under Audio, the Playback and Recording settings, and choosing a driver mode of ASIO? Bob Bone

I am on a tablet, at the moment, and will hunt up the support link, for you, in a few minutes. I have no such issue with Kontakt. 1) Please list the Kontakt version/build you are running. (Check for updates to Kontakt in Native Access. 2) Does this issue happen for any other loaded instrument, in Kontakt? 3) If there is some issue, you should be able to use Kontakt's Batch Functions, in its Output Section, to clear the output channels and assign a stereo output channel for each loaded instrument in the current Kontakt instance. You can then rename each output channel in the Kontakt Output Section, which should then clean up what you see when assigning audio track inputs to the corresponding loaded instruments in the Kontakt instance. Bob Bone

I have ThreadSchedulingModel=2, which is the middle value, and that works well, for me. In Preferences > Audio> Sync and Caching, I also have both the Playback and the Record I/O Buffer sizes set to 512 KB. @bgewin, have you evaluated performance with the loaded plugins temporarily bypassed? (hitting 'E' on your computer keyboard toggles effects bypassing On/Off) Do you have your Cakewalk Projects folder and any sample library folders Excluded/Exempted from your antivirus software? (if not, I suggest you do that). Bob Bone

Sorry, had your post up, but fell asleep. Certain types of plugins really put the whammy on things, if the ASIO Buffer Size (it is usually represented by some number of samples, like 32, 64, 128, etc., up to either 1024 or 2048), is set to anything less than 1024. These plugins that require such a large ASIO Buffer Size are designed to be used only during Mixing/Mastering, and not during the recording phase of a project's work flow. If your project has plugins loaded, such as a Convolution Reverb (Breverb, for example), or that use Linear Phase (LP64, for example), you will not be able to effectively be able to record, because they really require a super large ASIO Buffer Size to do their 'Look-Ahead Processing', where they read a large block of data ahead of the current position in the audio stream, and by looking ahead at the data that is coming, they then react in their processing. Anyways, a long while back, I learned that, for the rest of time, I needed to switch between a small ASIO Buffer Size (I use either 128 or 64), and a giant size (1024 or 2048), depending on if I am recording, or just doing playback during mixing. The smaller buffer, during recording, keeps the latency down low enough to where when I play notes, or sing, or whatever, it syncs up just fine to the other tracks. During recording, I ALWAYS make sure that if I have any of those plugins that require a giant buffer, I either power off those plugins, or swap them out for other similar kinds of plugins that don't require such a large buffer, and then I have no latency/processing issues because of those plugins. One other way of temporarily disengaging plugins, is to hit the letter 'E' on the computer keyboard, which is a Cakewalk shortcut key that toggles the bypassing of all effects in the project either Off or On. Any of those 3 methods will keep those plugins from interfering with the recording process. Conversely, whenever I am finished recording, and have moved on to mixing/mastering, I am now really in playback mode, so even though I then jack up the ASIO Buffer Size to either 1024 or 2048, that extra latency doesn't matter, because I don't have to try to sync up anymore, since I am not recording. If it takes an extra half-second before playback starts, it doesn't matter. The giant ASIO Buffer Size then allows those plugins that expect/require a large buffer, to work properly, and the universe becomes creamy smooth again. No matter which DAW, the combination of what plugins are active, plus the ASIO Buffer Size, and whether recording or just playing back, all have to work together, to avoid issues with streaming audio processing. Lots of times, the descriptions for a given plugin, will mention if it is designed for mixing/mastering, or if it uses 'look-ahead processing', and over time, you can learn and remamber which ones you need to make sure aren't active when recording. Other than that - just also remember to use a small buffer when recording, and a giant buffer when mixing/mastering or even just playback, when those types of effects are active. It is useful to look up the latency for plugins where you do not know how it will affect latency in you projects. Here is a link for Waves plugins, that has the latency of each plugin listed: https://www.waves.com/support/tech-specs/plugin-latency You can also do a Google search using a plugin's name and add the search term: 'latency', and you will frequently be able to see whether or not a given plugin should be only added or made active, once you have moved past recording and on to mixing/mastering. From the plugins you listed above, the GW Mixcentric, and the 2 Fabfilter plugins, all look like you shouldn't use them unless you have set your ASIO Buffer Size to a really large value, and that, of course, then means they are NOT meant for use during recording. Try your test again, but first make sure your ASIO Buffer Size is properly set to accomodate the specific active plugins in the project. Bob Bone

This only will help you moving forward - what I do, every time I am going to work on a project, is to create a backup copy of the project file (*.cwp). If I am going to be editing actual audio clips in the project, where I am going to do destructive edits - like doing things to audio clips, then my pre-session backup is of the entire project folder. Doing either of the above, as appropriate for the session, gives me a safe way to get back to where I was, so that the project is recoverable in a pristine state, should I accidentally do a boo boo and delete something accidentally, or I were to decide that I should not have deleted or modified something in a way where that was destructive - meaning it could not be undone. When I get to where the newest version of the project is indeed correct, completely, I will go back and delete older backup versions of the project - posibly leaving one prior backed up version from the current, just in case. I do the above manually, so that I have complete control over the backup process and the management of backup versions. When backing up, my naming of either the project file or the project folder, includes the date and usually some additional short description - at the end of the name - so I can see those versions in choronological and alphabetical order. Bob Bone

Try CTRL+F

I'm curious - could any of this involve the Stereo Panning Law setting, from Preferences? The mentioning of '3 db', in the above posts, made me wonder, so I thought I would ask. (I changed the Stereo Panning Law settings eons ago, from the default, to the -3dB center sin/cos taper, constant power setting, from some long ago tip from @Craig Anderton). (and, just as general observation - I found some of the help doc on the Panning Law settings, a bit murky - for example, it indicates that the panning law settings are applied once for tracks, but it does not say when it actually does apply it ). Here is some of the text from the Configuring Panning Laws doc: "In the Stereo Panning Law field, choose one of these options: (Default) 0dB center, sin/cos taper, constant power. This choice causes a 3 dB boost in a signal that’s panned hard left or right, and no dip in output level in either channel when the signal is center panned. -3dB center, sin/cos taper, constant power. This choice causes no boost in a signal that’s panned hard left or right, and 3 dB dip in output level in either channel when the signal is center panned. 0dB center, square-root taper, constant power. This choice causes a 3 dB boost in a signal that’s panned hard left or right, and no dip in output level in either channel when the signal is center panned. -3dB center, square root taper, constant power. This choice causes no boost in a signal that’s panned hard left or right, and 3 dB dip in output level in either channel when the signal is center panned. -6dB center, linear taper. This choice causes no boost in a signal that’s panned hard left or right, and 6 dB dip in output level in either channel when the signal is center panned. 0 dB center, balance control. This choice causes no boost in a signal that’s panned hard left or right, and no dip in output level in either channel when the signal is center panned. Pan Law compatibility mode When using a non-default pan law with floating point or 24-bit audio, Cakewalk would previously apply the pan law twice; once at the clip level and once more at the track level. In SONAR 8.5.2 and later, pan laws are only applied once at the track level and only for mono tracks. Any clip pan envelopes will continue to work, but behave strictly as a balance control. If you have existing projects that use a non-default pan law (i.e. other than 0dB center sin/cos taper), the mix might sound louder in Cakewalk. To address backwards compatibility with projects that were mixed in previous versions of SONAR, the following Aud.ini variable is available to set the pan law compatibility mode: PanLawCompatMode=<0 or 1> (default=0) This variable should be set in the [Wave] section. For example: [Wave] PanLawCompatMode=1 When the value is 0 (default), non-default pan laws are not applied at the clip level. Clip pan envelopes always use the 0dB center sin/cos taper law. When the value is 1, pan laws are applied at the clip level. It is not recommended that you use this value unless you need to retain backwards compatibility with pre-SONAR projects that use a non-default pan law."

Yup, I appreciated that post a bunch, too.

Did you mean that reverb was the only plugin loaded anywhere in the project?

The stuff that comes with Cakewalk, or prior Sonar releases, all work OK in 64-bit land. It is most likely some 3rd-party 32-bit effects or synths that I was referring to. Even if you cannot get to the testing I had asked about, can you perhaps list off every 32-bit plugin you have loaded into one of these projects where the audio droouts are occurring? I could then try to test them out on my system and see if I can determine if there are any issues when using them. Thanks, Bob Bone

It does seem like something may be going on, but that is only based on now several folks suddenly having issues with 32-but plugins in projects that just recently seeing dropouts. @kevmsmith81, can you try saving a copy of one of those projects under a new name (you can save the project in the same project folder, this is just a temporary thing to test without risk to your real project). So, once you save the project, make sure you have the newly-saved one open, and if the 32-bit plugins are synths, just swap them out for any 64-bit synth, and see if the dropouts stop. If the plugins are effects, then first try testing with those effects simply powered off, and if still dropouts, then try removing them, and see if that gets rid of the dropouts, again ONLY work in the newly created version of the project, so you don't accidentally do something to your 'real' version of the project. Also, please detail ALL 32-bit plugins that are loaded into the project experiencing these dropouts. Perhaps there is something some of us can try testing, using those specific plugins. Thanks, Bob Bone

Thanks, am looking for an amp modeling plugin for a guitar player, but it seems like more to it than feasible, because of the nature of 12 string high 2 strings not being doubled at the octave level, but are simply doubled B and E. I think this will need a real 12 string or samples, to meet my needs. Bob Bone

Yup - that was the 1st option I suggested, renaming that VST32 folder temporarily, until the corrected VstScan program is brought in, with the next update. That would give instant resolution to the issues - option 2 in my earlier post, is also viable, OP's choice. Bob Bone

Howdy - does anyone know of a good amp modeling plugin sound of an electric 12-string? I am trying to match the sound of Mike Rutherford's double-neck electric 12-string sound, as heard in Genesis songs like: Firth of Fifth, Squonk, Cinema Show, and lots of others. Bob Bone

Just a thought - I just checked on my system, and the B3 V2 comes in both a VST3 and a VST2 version, from Arturia. By default, Cakewalk will present the VST3 version. Some other plugins, in the past, have been a bit flaky in their VST3 versions, while the VST2 versions worked perfectly. SO - I suggest you try going into Preferences, and removing the check from the paramater in VST Settings where Cakewalk 'hides' VST2 versions of plugins when there are also VST3 versions available. This should then allow you to load up the VST2 version of the B3 V2 plugin, to see if that makes any difference. Please consider doing the above, as a test, and post back your results. Thanks, Bob Bone

Howdy - so, I inserted that exact same Redtron_SE plugin, into a new project, and it played just fine, so whatever is going on, it would seem to be with your system. Please tell me what version and build your Windows is - is it completely current on maintenance? How about the drivers for your audio interface? Up to date on maintenance? I think if we examine your configuration, maintenance levels, and possibly settings, we might figure out why this same plugin weirds out on your system, while playing just fine on mine. Bob Bone