You are using an out of date browser. It may not display this or other websites correctly.
You should upgrade or use an alternative browser.
You should upgrade or use an alternative browser.
OBS does not recognize all sound card inputs
- Thread starter Lorenz75
- Start date
You don't. At least not in OBS directly. It was made originally for the stereotypical bedroom streamer that only has two sources - a single mic, and a surround-sound game - each of which are different dedicated devices. So to avoid confusing him, you choose a device in OBS, and not the channels of that device. All channels are downmixed according to the surround standard that corresponds to that device's channel count. No way to change that. At least not natively.
You might try the ASIO plugin. It gives you a new source that lets you choose channels. But people have also reported that it doesn't work with the current version of OBS. Might still be worth a shot though.
Or, you could move ALL of your audio work out of OBS and into a DAW (Digital Audio Workstation). Lots of those to choose from, with varying prices - sometimes free - and varying functionality, and those two things don't seem to be correlated anymore.
Then everything goes directly to the DAW, not OBS, and all the processing happens in the DAW, not OBS. OBS knows nothing about audio at all anymore, except to receive the final soundtrack from the DAW, as its *only* audio source, to pass through completely unchanged.
You might try the ASIO plugin. It gives you a new source that lets you choose channels. But people have also reported that it doesn't work with the current version of OBS. Might still be worth a shot though.
Or, you could move ALL of your audio work out of OBS and into a DAW (Digital Audio Workstation). Lots of those to choose from, with varying prices - sometimes free - and varying functionality, and those two things don't seem to be correlated anymore.
Then everything goes directly to the DAW, not OBS, and all the processing happens in the DAW, not OBS. OBS knows nothing about audio at all anymore, except to receive the final soundtrack from the DAW, as its *only* audio source, to pass through completely unchanged.
Ciao, grazie per la risposta, ma non volevo usare una DAW, per non gravare sul processore del PC, in quanto i 7 sono di quarta generazione, quindi molto vecchi. Pensavo di utilizzare OBS e basta, perché in alcune interviste ho la necessità di utilizzare più di due microfoni a condensatore, per cui la Scarlet 18i20 ha la possibilità di collegare fino a 8 microfoni con i 48 volt e sfruttare anche gli altri ingressi di linea. Attualmente i microfoni sono collegati ad un mixer Yamaha mg10 xuf, il quale a sua volta è collegato alla scheda audio nei primi due canali della stessa. il problema è che nel mixer i 48 volt sono universali per tutti e 4 gli ingressi di linea, e non è possibile scegliere il canale da inserire con il microfono che non necessita di questa alimentazione. Come si potrebbe risolvere questo problema?Tu no. Almeno non direttamente in OBS. È stato realizzato originariamente per lo stereotipo dello streamer da camera da letto che ha solo due sorgenti: un singolo microfono e un gioco con audio surround, ognuna delle quali è un dispositivo dedicato diverso. Quindi, per evitare di confonderlo, scegli un dispositivo in OBS e non i canali di quel dispositivo. Tutti i canali vengono sottoposti a downmix in base allo standard surround che corrisponde al numero di canali di quel dispositivo. Non c'è modo di cambiarlo. Almeno non in modo nativo.
Potresti provare il plugin ASIO. Ti offre una nuova fonte che ti consente di scegliere i canali. Ma le persone hanno anche riferito che non funziona con la versione attuale di OBS. Potrebbe comunque valere la pena provarci.
Oppure potresti spostare TUTTO il tuo lavoro audio da OBS e in una DAW (Digital Audio Workstation). Ce ne sono molti tra cui scegliere, con prezzi variabili - a volte gratuiti - e funzionalità variabili, e queste due cose non sembrano più essere correlate.
Quindi tutto va direttamente nella DAW, non in OBS, e tutta l'elaborazione avviene nella DAW, non in OBS. OBS non sa più nulla dell'audio, tranne che per ricevere la colonna sonora finale dalla DAW, come *unica* fonte audio, per passare completamente invariata.
It's rare and expensive to have a mic that is actually damaged by phantom power. They do exist, but if you have one, you probably know. The vast majority of mics that don't need it, are just fine to have it anyway. They just don't draw power from it and continue to work the same way.
What's far more common, is to have something like a direct connection to a computer, phone, or similar thing, without a DI box, that can't handle the 7mA per wire that +48V phantom power delivers while the device tries to hold its output at 0V. A DI transformer will happily sit at +48V, drawing nothing, as will a mic that doesn't use it.
If you're otherwise happy with the physical console, keep using that, and don't worry about having phantom power on mics that don't need it. They're fine. Just use a 2-channel interface from the board to the PC, instead of a multichannel one that you only want the first two of, because of how OBS handles multichannel devices.
---
If you want to use a DAW instead, that's okay too. Compared to processing live HD video, audio is hardly anything. So I wouldn't worry about that.
For some numbers, 48kHz audio with 30fps video, gives you 1600 samples per frame. Assuming 4 colors per pixel (RGB + transparency), with 8 bits each, that's 32 bits per pixel. 32-bit audio samples are also common as an internal format, though it's towards the low end of things. (OBS uses 32-bit audio internally) 64-bit is more common for serious audio work. That's vastly overkill to actually listen to (16-bit done right, is already indistinguishable from analog), but it means that the developers don't have to worry at all about the roundoff error building up to something even remotely noticeable.
So, if we have 1600 samples per frame, and each sample is upconverted to 64 bits in the DAW, that's equivalent to 2 pixels per sample. A 1920x1080 frame has 2,073,600 pixels. Divide that by the 2 pixels per sample and 1600 samples per frame, and we're left with a factor of 648. You need 648 channels of 64-bit 48kHz audio to equal the data rate of a single video source at 1920x1080p30.
Even if you count each intermediate stage of audio processing as its own channel, which might not be a bad idea, it still takes a LOT to get 648 of them! So, compared to processing live HD video, audio is hardly anything. I wouldn't worry about that.
What's far more common, is to have something like a direct connection to a computer, phone, or similar thing, without a DI box, that can't handle the 7mA per wire that +48V phantom power delivers while the device tries to hold its output at 0V. A DI transformer will happily sit at +48V, drawing nothing, as will a mic that doesn't use it.
If you're otherwise happy with the physical console, keep using that, and don't worry about having phantom power on mics that don't need it. They're fine. Just use a 2-channel interface from the board to the PC, instead of a multichannel one that you only want the first two of, because of how OBS handles multichannel devices.
---
If you want to use a DAW instead, that's okay too. Compared to processing live HD video, audio is hardly anything. So I wouldn't worry about that.
For some numbers, 48kHz audio with 30fps video, gives you 1600 samples per frame. Assuming 4 colors per pixel (RGB + transparency), with 8 bits each, that's 32 bits per pixel. 32-bit audio samples are also common as an internal format, though it's towards the low end of things. (OBS uses 32-bit audio internally) 64-bit is more common for serious audio work. That's vastly overkill to actually listen to (16-bit done right, is already indistinguishable from analog), but it means that the developers don't have to worry at all about the roundoff error building up to something even remotely noticeable.
So, if we have 1600 samples per frame, and each sample is upconverted to 64 bits in the DAW, that's equivalent to 2 pixels per sample. A 1920x1080 frame has 2,073,600 pixels. Divide that by the 2 pixels per sample and 1600 samples per frame, and we're left with a factor of 648. You need 648 channels of 64-bit 48kHz audio to equal the data rate of a single video source at 1920x1080p30.
Even if you count each intermediate stage of audio processing as its own channel, which might not be a bad idea, it still takes a LOT to get 648 of them! So, compared to processing live HD video, audio is hardly anything. I wouldn't worry about that.
Last edited: