By the way I’ve recently read about this thing. What a scam! No proper DAW interface, no VST compatibility, no built in reverb! It can’t save the last tune, and needs re-tuning regularly. The user interface is complicated and terrible to learn. Obviously not suitable for making proper music.
Looks like a demo or toy that exists within a bubble.
(Warning! This post contains significant amount of sarcasm.)
Indeed, I have the pro version and never use it… For me VCV is what helped me explore new ways to create music and sound, not just a fancy synth / effect. Also, recording inside VCV and importing in whatever (free I guess ?) DAW you want to use is absolutely valid, works well and has been a good creating technic since the 60’s. We could go on trolling with an incredibly long list of demo / toys that have the same problem that VCV Free has, that would make for a good fun moment ! TB 303 ? TR 808 ? Organs ? Cowbells !!!
In the end, if you want a good sounding modular synth, you have even free ones that work really well. If you want a modular approach you don’t need the DAW integration, if you want to actually just make music, you don’t need either !! If you only want to complain, you need EVERYTHING AND NOTHING AT THE SAME TIME.
Have fun with what you have, be creative !
I use the standalone all the time, never use the VST, but I’m not much of a plugin-oriented musician anyway. Also, I compile VCV Rack and most of its plugins myself. It’s all free, the Rack code, the plugins code, and the tools to compile it. Building software from source code isn’t rocket science. I’m a musician by training and trade, but even I can do:
./configure
make
sudo make install
(An oversimplification of the process, but you get the point).
It’s also worth noting that there’s a vast amount of very fine music software available for free.
If you absolutely must have a VST version look at Cardinal (already suggested, I know).
And btw, you are welcome here. 
Best regards,
dp
Agree++. However:
Free drums I can get from the Befaco Kickall, Vult’s Trummor, Jerry Sievert’s DrumKit, and the Autodafe drums are still working on v2. And of course there are the myriad percussion sample sets for use in the Rack samplers.
Convolution reverb I can get from Csound via the dbRackCsound plugin or from NYSTHI’s Convolvzilla. Chorus is available from Surge XT, Rackwindows, Chowdsp, and others.
I’ve been using Csound on & off since 1989, the docb modules really elevate the Rack experience for me.
Best regards,
dp
Yeah! My point exactly!
BTW, I am very bad with Csound, but I know you can do crazy stuff with it. My personal record - making a sine wave there. After that I realised that I am not very smart, hahaha. But I know that you can just google it and copy-paste. I have a question though: is dbrackCsound easier on CPU when compared to Convolvzilla (cause it just burns all of my free CPU for some reason). I love the IR stuff and I was using it not very often, because of that. But if Csound works lighter than Convolvzilla, that’s an absolute win
I was very excited to get VCV pro, after a year of saving 
. But now that I have it, I find it difficult to switch back into a DAW environment to compose. It’s probably just a mindset, but yes, Rack standalone is still tops. I knew nothing about modular synthesis when I found Rack, and when I figured it out, my world just exploded. This would not have happened if Rack was not free in the first place.
I’ve gone down two paths, depending on how I feel about things.
Build up entire tracks/pieces/songs in Rack. I have a method for this: use 8Face to save snapshots of sometimes large groups of modules, then use channel mutes in Mindmeld Mixmaster to bring sounds in and out. That’s pretty much Detroit techno compositional technique from 1991, but it’s surprising how much variety, structure and intentionality it allows for. And actually there’s nothing wrong with Detroit techno circa 1991, heads still aint ready.
Use Rack VST in Ableton Live, but often in MIDI-only mode. I can then use all the absurd variety of VST instruments and effects (and the Ableton Live Suite instruments & effects) but using the sequencers in Rack to control them. I will also use the Live FX VST for effect sends, to do some things that come natural in Rack but are difficult or awkward in Live.
Given that I literally have ALBUMS worth of material mostly done with method #1, I would not give up on this method. It may take you a while to find your workflow but the components are all there. And I use the paid plugins a LOT, but there’s so much that is possible with JUST the free plugins, they’re optional.
I’ve bought a lot of them, as simply the latest iteration of my gear acquisition syndrome, but my real favorite, constantly used modules are all open source and free. I don’t regret buying them because spending money on paid modules supports the Rack ecosystem, but if they went away tomorrow, I’d still be able to do the things I want to do.
TL;DR embrace the free; it is not limiting, it’s empowering, particularly for people on fixed incomes.
I think it is fair to say that Rack-free with the paid HOST and Rack pro with VST+ capabilities are very comparable in terms of capability. I have worked exclusively in DAWs (Ableton Live most recently) over the years. But all of my original experience with modular was in hardware 50 years ago this year. So, I like the virtual modular stand-alone environment, as long as I can get to my my VST plugins via HOST.
It comes down to personal preference and budget is a reality.
I have the rack plugin open on one monitor, the DAW on another. It’s no different to working in standalone rack to me.
I ran a simple test with a 1.1 MB impulse response file in a CsoundFX module (mono) and in Convolvzilla. The scripting module used example code from the online reference manual entry for the pconvolve opcode. At the default settings the performance meters reported 10% in the Csound module with regular spikes at 20% (related to a partition size parameter; no audio problem though). Convolvzilla reported a mere 0.7% with no spikes. Of course, the scripting module is likely to suffer a performance hit, but the NYSTHI module appears to win big for realtime purposes.
However, the performance stats inspired a closer look at the Csound code. I altered that partition value from 1024 to 512, but the performance hit doubled. So I set it to 2048 and the meter now reports 0.4% with regular peaks at 14%. Not great, but bearable in the absence of the NYSTHI convolver.
Interesting test, thanks for the notion. The Csound script can likely be further optimized for performance, and I think some realtime controls can be added. Convolvzilla has a useful GUI, adds predelay and early reflection parameters, and is lightweight here (Mac Mini M1, Rack with Rosetta). Both are free, nice to have both.
Apologies to the OP for the threadjacking.
Best regards,
dp
A lot of plugins that do block convolution (like my Colors) use a worker thread to do the big FFT. That keeps those spikes off the audio thread, and makes them not show up in VCV’s meters.
As someone pointed out very astutely here recently, this works great if no one else does it. If a lot of modules do it at the same time then it’s a free for all and bad things can happen.
I have seen modules that do the FFT on the main auto thread, too. This of course makes them “spikey”, and can lead to pops and clicks.