Monitoring, latency, ASIO, etc

[Dazzer1234]

Hi folks,

I'm trying to set up OBS but i'm struggling with latency.

This is what I'm trying to achieve:

My capture sources are:

Full screen
DAW output
Microphone input

The DAW output and microphone input come into OBS via the hardware mixer on an RME Fireface UC.

The essential problem seems to be that because OBS can't use ASIO drivers (i'm on PC), i will never be able to monitor through OBS while having an acceptable amount of vocal latency in my headphones.

I'm basically using OBS to capture tutorials which shouldn't have any editing afterwards, so i want to get the sound right at the time of recording. Ideally this would mean monitoring through VST plugins for compression / EQ while recording. But without ASIO, this isn't possible while having acceptable latency.

I'm wondering how others tackle it? It must be a common problem for streamers, right? You want to know that your sound is good as you hear it in your headphones while streaming, and the only way to be sure of this is to be monitoring the output of your recording device, which is OBS.

Cheers and thanks in advance!

 

[Lawrence_SoCal]

Hopefully AaronD will chime in

My understanding is that best practice is to mix outside of OBS Studio (and monitor, as you have observed). Then you have a one-time config of the input from your external mix(er) into OBS. And typically, that won't change over time.
Then there is the issue that almost all 'free' CDN (content delivery networks) re-encode both audio and video (heavily compressing), so no matter how good it sounds in OBS Studio, what you viewers hear may be slightly different. There are hacks, temporary work-arounds, etc... but no real fix to CDN re-encoding (other than using a pay service with higher quality... but that really is a different market than you are probably talking about)

do NOT take my word for this... but I seem to recall that *if* you do ALL audio mixing outside OBS Studio, And then do nothing but pass-thru the audio in OBS Studio (no compression/limiters/effects/plug-ins, etc) then you could Output and Monitor in OBS Studio and hear what OBS Studio is processing. Again, iirc (and I may easily have this backwards), OBS Studio Monitoring is the Audio Input, not the post-processed Audio Output

To hear OBS Studio post-processed audio, I think you have to do some custom audio routing to hear that Audio Output as well as Recording/Streaming (maybe using virtual adapters, or Presentation mode, or ??) but then ... latency

I'm sticking with Windows, but my recollection is that is high-quality, and complex audio processing is important to you, you'll switch to MacOS or Linux