Yes, Voicemeeter is my usual suggestion over other alternatives because it's easy to configure. However, latency is always the gotcha when adding more software layers.

I haven't tried "potato" - although I admit not having used it in quite some time. I used standard Voicemeeter.

Looks like like each level you get more I/O channels. Voicemeeter standard as least, then Voicemeeter Banana as more, and Voicemeeter Potato as the most.

i am not sure a more powerfull PC will solve the buffer/latency issue using multiple engine with Voicemeeter, mine is not so bad already; I7 4790k over 4ghz and 32 gb ram. so i came back to the old world, switching audiocard up to my need, i keep the voicemeeter when i want to check all sound together from pc, ipad and montage.

Your CPU is 7 generations old. Later gen CPUs can probably bump up the single-thread performance and help you out. To what degree - I don't know - that's not my area as I'm not a PC/CPU/plugin/etc. user.

Here's an interesting thread from several years ago discussing latency. The specifics (CPU models) are no longer current - but general themes are still worth looking at.

https://vi-control.net/community/threads/pc-specs-for-lowest-possible-latency.78570/

Thank you Jason

A new software is here for Mac and coming for Windows. https://www.gingeraudio.com/
I haven't used it but thought others should know about it. Just do your research before installing - its very new.

Sorry but I'm hardly figure out what this thread has to do in Montage chapter ?
Is there any link ?

Fred,

This (and several other) thread(s) deal with using Montage with a computer and routing audio within the computer. Montage is involved because it is either the target or source of digital audio through the USB TO HOST connection on Montage to a computer's USB port.

One of the challenges of routing audio within a computer is that the operating system, driver, and software utilities work within a paradigm that there can be only a single audio interface. If you choose to use Montage for its multitude of stereo digital outputs (one stereo output per Part) - then this means you are "locked out" of using any other audio interface. Particularly in the Windows PC environment as the operating system itself does not have something available in Macs called "Aggregate Devices".

In order to overcome this limitation (not a Montage limitation, mind you - but an operating/computer imposed one) - there are some "hacks" one can do in the Windows environment or, if in the Mac environment, using Aggregate Devices - or perhaps another "hack" of using 3rd party solution as Chris has linked to.

The thread is similar to any other which involves software interfacing to the Montage. It's not all about the keyboard itself - but sometimes how to make the keyboard work within the environment of other devices (computers, other MIDI devices, sound reinforcement, etc).

I love it ... besides being extremely latent... it works great! lol...

Even when you have multiple audio sources (like setting up a Aggregate Audio Device in a Mac) you still can only connect one of the audio interfaces to your speakers at a time. Therefore, one of them must becomes your principal Audio Interface (it will be the one that allows you to Monitor yourself *direct* - Direct Monitoring is not “low latency” — it is No Latency) given a choice between “low latency” and “no latency” it is no contest which is preferred.

No latency direct monitoring allows the musician to play without concern, and is always preferred — the DAW will time stamp, and place the recorded signal exactly where it belongs. You will need this the more simultaneous inputs you open. No latency means you are listening before the audio signal goes to the computer. You are listening to yourself “live” — literally, direct monitoring. Audio latency occurs when you are listening to audio that has travel via USB to the computer, where it is time stamped, and placed on your computer’s hard drive, before heading back to the audio interface for D-to-A conversion, so you can monitor it in your speakers. This takes several (precious) milliseconds and is what is called “latency”.

So for the same reason when recording to old style analog tape machines, the recording engineer monitors the band live; the signal sent to the player’s headphones is the *direct* signal... you don’t listen to the tape until you playback. The record head is before the playback head. And therefore it is like the “latent” audio signal. You monitor signal before it is recorded to avoid the latency entirely.

The direct monitor is “speed of light” — no latency.
Monitoring yourself post (after) the computer — latency applies.
The more channels simultaneously open, the more CPU muscle required.
(This is why, in a one-person-operation, it just simply is overkill).
Direct Monitoring (no latency) is better than even single digit millisecond latency. You can avoid it entirely. Monitor yourself direct! Whenever possible... if you’re doing some effect processing to your signal via plugin effects — obviously, if want to hear those effect while you are recording, well you must monitor the latent signal. Oh, and by the way, latency increases as the computer CPU must generate and apply the plugin effects.

No magic bullet—
This means one of the Audio Devices must receive analog audio signal from all other audio sources so that you can monitor them (in real-time) with “no latency”. There is no magic... the more audio inputs that are simultaneously active, the larger the Buffer Size you’re going to need to accommodate handling all the inputs. And if you don’t have direct monitoring, you must operate at the business end of the latent signal.

If your only goal is to avoid learning how to reconfigure your gear efficiently or you are just too lazy to unplug and replug your gear for what you need to do, remember: the computer is going to work best when you are setup to record, efficiently. If you have a home studio and it is just you... setting up to record 48 audio inputs at once, is, well, just silly (the return on investment just doesn’t make sense).

Know that if you record your MONTAGE as your basic music tracks.... once you render them as audio, you can reconfigure your system to use a different audio interface to overdub your vocals (maybe you have a fancy boutique phantom powered audio interface for your large diagram condenser microphone). Your phantom powered boutique audio interface will allow you to playback MONTAGE audio Tracks while recording your microphone overdubs.

If you don’t need to play and sing simultaneously, if you don’t have sessions where you need to record several musicians live, simultaneously, it’s going to be more efficient to have one audio configuration for your music tracks... then a different configuration for laying down your mic’d overdubs. Yes you can use various audio interfaces on the same project. Just select a single audio resolution to work at so you don’t wind up asking the computer to juggle different sample rates.

Setting up an Aggregate Audio Device (or coming up with any scheme to have more than one audio interface active)... say you configure all 32 MONTAGE Inputs plus the 8 or 16 audio inputs of your boutique audio interface for your microphones when you are not even going to record them simultaneously... is like, really not efficient.

Word to the wise... if it is just you recording alone in your home studio... multiple (aggregate) audio device configuring should be considered only after some serious thought (please)... just my 2 cents... before you throw your money at the issue, evaluate what it is going to accomplish

...if this whole “direct monitoring” situation is still unclear to you, please learn about it before throwing money down this rabbit hole.