msmcleod

This should give you some idea of the speed increase. It looks like the new CPU will be between twice as fast, to 2.5 x faster with 8 cores: https://cpu.userbenchmark.com/Compare/Intel-Core-i7-11700KF-vs-Intel-Core-i5-3470/m1514594vs2771

It's mostly about choosing the right algorithm. I find polyphonic sustain works best for me for most cases. For Melodyne 4, this means creating a region fx, setting the default algorithm, then removing the region fx before you drag to the MIDI track. If you're using Melodyne 5, you'll get a prompt to choose the algorithm before the conversion. Sometimes applying a high & low pass filter to the source, and/or adding some compression can help too. You'll need to bounce the result so it's baked into the waveform before converting to MIDI. This reduces the frequencies Melodyne has to analyse (in particular high frequencies / harmonic content), and evens out the velocities. Mess around with normalization as well to get a consistent signal level for Melodyne to use. You can get reliable results pretty quickly with this method, but it can take time to get good results with respect to velocities. Once you get some settings that work well, save it as a preset (i.e. in an FX chain) so you can go to it quickly. Alternatively, you could use a CAL script or Articulation Map to even out velocities afterwards. Velocity values are always going to be a problem, as they're not only sensitive to the source material, they're also gonna be different depending on the instrument you're playing the MIDI back with.

I've used this in the past - the tracking and recognition are least as good as my old Roland GI-10 with a GK2-A pickup ( which to be fair, isn't that great either), but the fact that it does it with a single audio signal and in realtime is impressive. However... the plugin itself is pretty buggy in my experience, and very sensitive to both ASIO buffer size and sample rate. My best results were at 48Khz with a buffer size of 128 or 256. When I did use it, I bounced my existing project as a single WAV file (as a backing track), and imported it into a brand new project for doing the MIDIGuitar stuff. That way, if it crashed (which it did quite often), I wasn't worried about corrupting my existing project. I've since stopped using it to be honest. I can get just as good results just recording the guitar as audio, then dragging the audio to a MIDI track and making the required edits for note velocity. @Kurre - this is what I'd recommend: 1. Drag your recorded guitar audio on to a MIDI track to convert it to MIDI 2. Insert Cakewalk's Chord analyser on the MIDI track as @57Gregy suggests.

I gave up on using computers live a long time ago. Going from a cold car to below zero outside then straight into a pub with a blazing fire going, then the reverse at the end of the night soon takes it's toll on a PC, and its a disaster if it fails or glitches during a performance. I use Cantable as a host to my VSTi's, then run SampleRobot to sample the VSTi. I then put the samples as sf2 files on to one of these, running SampleLord: While it can actually run VSTi's natively, it's way underpowered, so I find sampling my synths and using SampleLord on the V-Machine far better for CPU, and also more convenient as I can have a single sf2 with a whole song's worth of synths in it as separate internal patches. I'll maybe run dexed or Cobalt natively on the V-Machine for sounds that need extra realtime control.

For plugins, CPU performance is most important. For projects with lots of sample libraries, RAM is important. So ideally you want both. As for a performance improvement, I've used to have an i5 3570 processor which I upgraded to an i7 3770, which is comparable to your setup given it's a 3rd gen CPU. Both are running at 3.4Ghz. I saw maybe a 20% - 30% improvement going from the i5 to the i7 - not huge, but then the speed is the same, and although the i7 is hyperthreaded, there's still only 4 cores. My 6th gen i7 laptop @ 2.6Ghz matches the performance of my 3rd gen 3.5Ghz i7 3770. My i7 - 9750H @2.6Hgz laptop is significantly faster than the other machines ( at least twice as fast ), so a ~4GHz 11th gen i7 should be even faster.

Actually, @Cea - could you send your Cakewalk project file (.cwp file) to @Jonathan Sasor ? We've had another report of a crash on project open today, so it would be good to know if this is the same issue. Thanks.

@Cea - you can rollback to 2021.04 using this link: Cakewalk 2021.04 Rollback installer.

I'm not sure this is the fault of PC-2A... it's a different plugin to CA-2A. I've got both on my system, and I don't get this. It's more likely your CA-2A installation has been corrupted somehow, or maybe you moved it or have duplicate copies ? I would suggest re-installing CA-2A.

The export dialog is selection sensitive, and the source category specifies where it captures the source. Selecting a source category of "Entire mix" with a single clip or track selected, means put that single clip or track through the entire mix - i.e. through all the buses etc, and capture it after the master bus. Selecting a source category of "clips" will capture the selected clips as individual files before it hits the track - i.e. there will be no track fx, no pro channel, and no track volume or panning. Selecting a source category of "tracks" will capture the selected tracks as individual files before it hits any buses or the master bus. I quite often want to capture my tracks as it sounds through the master bus - not how it sounds before it hits any buses. In this case I'll select the track prior to opening the export dialog, and select "Entire Mix". Alternatively, I'll not select anything, but solo the track and select "Entire Mix"

Offset Tempo map works on the node selection - so make sure you have them all selected. You may find them easier to select from the tempo inspector.

VST3's should only ever live in C:\Program Files\Common Files\VST3 , and there should only be VST3 files in that directory. SF2 files are not virtual instruments, they are multi-samples in SoundFont format. You need an SF2 player to use them (such as Sforzando, TX816W etc)

If you're using a Workspace, save it after re-ordering your control bar modules. Don't save over the factory ones though - give it a new name else it'll be overwritten when you next update.

This was useful in the early 90's when people were migrating from older sequencers. For example, I used Music-X on the Amiga before moving to Cakewalk which had a resolution of 192 ticks per quarter note. I spent most of my time in Music-X's event view, so being able to set Cakewalk's ticks per quarter note to 192 made the transition much smoother when using Cakewalk's Event View., until I slowly changed my workflow to use Cakewalk's other features.

You can change the time base to Absolute time on an arranger track by track basis. So all sections within an absolute time track will keep their absolute time position and length regardless of tempo changes. To change the time base of an arranger track, either use the arranger track context menu, or click the time base icon in the arranger track header:

You might want to try @scook's CbB Tools - in there is a utility to populate the Utilities menu. It's towards the bottom of this page: https://docs.google.com/document/d/1eiVH_I1clbbRtWFh4-3Mo7HANjCxR8SwqHJtYXy19gw/pub

When using the smart tool, you can use CTRL + SHIFT to stretch the clip to make it shorter:

@Noel Borthwick's fix was a generic one for both extra ports appearing, and ports being missing from updated VSTi's, so the fix in EA should equally apply to those as well.

+1 to this.. I had a brown out a few years ago, where the AC went down from 240V to 50V. Three SSD's were destroyed by that.

If I were to pick my top two synths, it would be Omnisphere and Synthmaster 2.9. Both synths have a learning curve, but both also have a wealth of presets, so it's not too difficult to take a preset close to what you want then tweak it. They both have multiple synthesis types. No.3 on my list is Cobalt, which is a free VSTi similar to the Korg DW8000 / Kawai K3.

There's also Sonivox Orchestral Companion Brass, which has solo brass instruments, including a french horn. It goes on sale quite often ( I think I paid about £5 for mine), but I can't see any decent deals ATM.

^^^ This . Switch your driver mode to WASAPI within preferences, and see if the problem goes away.: Note there's two modes of WASAPI: WASAPI Exclusive: lower latency, but Cakewalk will have exclusive control over your audio device (i.e. you'll no longer be able to listen to YouTube etc when Cakewalk is using it). If this is an issue for your workflow, consider checking "Suspend Audio Engine When Cakewalk is Not in Focus". Also, not all audio devices support this mode. WASAPI Shared: higher latency, but the audio device will be shared with all other applications. You'll want "Suspend Audio Engine When Cakewalk is Not in Focus" to be unchecked in this mode. You may find WASAPI Exclusive is more suitable for recording real-time MIDI, or recording guitar/vocals when monitoring through FX. WASAPI Shared should be fine for everything else.

IIRC, MidiGuitar is very sensitive to ASIO buffer size and sample rate. It's been a while since I've used it, but it needs a minimum of 128 as a buffer size and it worked best for me at 48K. Any time I used it in a project, I'd delete it as soon as I had the MIDI conversion. But yeah, I stopped using it due to it being buggy and prone to crashes.

It's probably easier to just zip up the folder with your project in it, so we have both the project & audio files, then send that zip file to @Jonathan Sasor

@Mark Bastable - can you send a project that crashes as a bundle file to @Jonathan Sasor , along with a note as to which track you were trying to bounce? We'll look into it.