Screen Saver won't recognize built-in mic audio: MacBook Pro

[ciphex]

Hey guys,

I'm trying to get Gforce (3.7.4) Screensaver to recognize the built-in mic input on my Penryn MacBook Pro running Leopard. The standalone seems to work just fine as does the V-bar when i set their options to use the internal mic.

I read the instructions on changing the input method on standalone and checking the prefs file for the A_in= value.
eg. ~/Library/Preferences/SoundSpectrum/G-Force/Preferences (Standalone).txt

...maybe i'm missing something but the value in that file is always A_in=3 no matter what i set the source to using the toolbar.
That value however gives me an error when i run the screensaver: "error opening the preferred audio input device."

The same for the V-Bar file. it doesn't change either. I have tried values that do not throw errors in the screensaver. -2, 0, 1 & 2 do not throw errors, but they do not pic up the mic signal either. other values throw the error i mentioned before.

How can i get my screensaver module to communicate properly with the internal mic?

/me wishes that apple would have left coreaudio more open

[juxtiphi]

Hey guys,

I'm trying to get Gforce (3.7.4) Screensaver to recognize the built-in mic input on my Penryn MacBook Pro running Leopard. The standalone seems to work just fine as does the V-bar when i set their options to use the internal mic.

I read the instructions on changing the input method on standalone and checking the prefs file for the A_in= value.
eg. ~/Library/Preferences/SoundSpectrum/G-Force/Preferences (Standalone).txt

...maybe i'm missing something but the value in that file is always A_in=3 no matter what i set the source to using the toolbar.
That value however gives me an error when i run the screensaver: "error opening the preferred audio input device."

The same for the V-Bar file. it doesn't change either. I have tried values that do not throw errors in the screensaver. -2, 0, 1 & 2 do not throw errors, but they do not pic up the mic signal either. other values throw the error i mentioned before.

How can i get my screensaver module to communicate properly with the internal mic?

/me wishes that apple would have left coreaudio more open

normally the screen saver does not react to music at all


the link below where this was discussed, read the whole thread. i tried the method suggested and could not get the SS to react to music and there is some mention of mac.

but why wouldnt you just use the standalone thats what its for

http://www.soundspectrum-forums.com/viewtopic.php?

[ciphex]

juxtiphi,

the link you shared does not go to a valid post... if there is more reading somewhere though I would be very glad to know of it.


Here is the thing to the best of my understanding after investigation:

The screensaver should operate as a module just like the standalone or V-Bar does... same code just a different container. the difference is that the toolbar app doesn't interface with it.


There are a few things that may be causing confusion here:
Under Leopard the OS has a different method of routing sound through the core-audio framework which prevents an application from grabbing the raw system output. Also on Intel Macs this is confounded by the fact that the input hardware presents itself independently of the output hardware, instead of appearing as an aggregate device. These things together prohibit and confuse apps that have previously worked well in OS X.

Now, I understand this limitation. So an application on my mac cannot, for this poorly conceived reason, address the output of the system audio device unless it either is itself the source of the audio or it implements itself as a driver (user-space or kext) - and even then it's tricky.

So to get around this I just want to get Gforce to listen to my built-in mic for a signal to process. This is supposed to work, i mean, the screensaver may default to no signal input (or an artificial signal) but i think it's pretty useless that way. Gforce was meant to have a real-world signal.

To answer your question: i have a goal here. I want to run Gforce visuals as my desktop background in certain circumstances so that it properly responds to the environment around me. So that the machine feels to breath the atmosphere where i am while still being functionally useable.

I have this all ready and set to go except for the input source. I cannot get it to read the signal of my mic. There must be something I am overlooking...

Andy, if you're reading this... been a long time man, Gforce is really awesome. It's been a pleasure to watch it progress. Sure has come a LONG way

If there's any way to get this functioning in Leopard, i will be happy to provide debug output of any kind in an effort to get it worked out.

[juxtiphi]

juxtiphi,

the link you shared does not go to a valid post... if there is more reading somewhere though I would be very glad to know of it.


Here is the thing to the best of my understanding after investigation:

The screensaver should operate as a module just like the standalone or V-Bar does... same code just a different container. the difference is that the toolbar app doesn't interface with it.


There are a few things that may be causing confusion here:
Under Leopard the OS has a different method of routing sound through the core-audio framework which prevents an application from grabbing the raw system output. Also on Intel Macs this is confounded by the fact that the input hardware presents itself independently of the output hardware, instead of appearing as an aggregate device. These things together prohibit and confuse apps that have previously worked well in OS X.

Now, I understand this limitation. So an application on my mac cannot, for this poorly conceived reason, address the output of the system audio device unless it either is itself the source of the audio or it implements itself as a driver (user-space or kext) - and even then it's tricky.

So to get around this I just want to get Gforce to listen to my built-in mic for a signal to process. This is supposed to work, i mean, the screensaver may default to no signal input (or an artificial signal) but i think it's pretty useless that way. Gforce was meant to have a real-world signal.

To answer your question: i have a goal here. I want to run Gforce visuals as my desktop background in certain circumstances so that it properly responds to the environment around me. So that the machine feels to breath the atmosphere where i am while still being functionally useable.

I have this all ready and set to go except for the input source. I cannot get it to read the signal of my mic. There must be something I am overlooking...

Andy, if you're reading this... been a long time man, Gforce is really awesome. It's been a pleasure to watch it progress. Sure has come a LONG way

If there's any way to get this functioning in Leopard, i will be happy to provide debug output of any kind in an effort to get it worked out.

Ok I see where you are going with this, you want GF to react to the ambient sounds of your livingroom. the next line is a quote from keycompton about mac os I dont know if it applies but,

Unfortunately the Mac OS does not allow G-Force to access the audio driver information -- however, if you have an iSight or any other microphone device, you will find that this method works very well.

on the other hand make sure that the overall volume of your mic is set to max and even then it may not be loud enough for GF to hear ( the lower the volume the less GF reacts)

so maybe if you were to put an amplifier between your mic and mac to make the signal stronger it may help.

[ciphex]

this is a laptop. the mic is built in, the sensitivity is great and works just fine in the other applications: V-Bar, Standalone. In fact it picks up my keystrokes from feet away. it responds even to heavy breathing.

the problem lies exactly at the point where the screensaver is just not recognizing the mic as an input source though the other applications do.

I am trying to get my screensaver prefs set correctly so that they listen to the mic input.

i can specify this using the toolbar for the other apps. but the toolbar doesn't recognize the screensaver module so it doesn't connect to it to set the prefs. so i have to try and edit the screensaver prefs manually. however the values I am giving the screensaver config file don't seem to work. so i am trying to get some technical idea of how the screensaver looks for valid/invalid inputs.

[juxtiphi]

this is a laptop. the mic is built in, the sensitivity is great and works just fine in the other applications: V-Bar, Standalone. In fact it picks up my keystrokes from feet away. it responds even to heavy breathing.

the problem lies exactly at the point where the screensaver is just not recognizing the mic as an input source though the other applications do.

I am trying to get my screensaver prefs set correctly so that they listen to the mic input.

i can specify this using the toolbar for the other apps. but the toolbar doesn't recognize the screensaver module so it doesn't connect to it to set the prefs. so i have to try and edit the screensaver prefs manually. however the values I am giving the screensaver config file don't seem to work. so i am trying to get some technical idea of how the screensaver looks for valid/invalid inputs.

with all the ways you have looked at it i am led to believe there is no way to get it to work for now.

[ciphex]

with all the ways you have looked at it i am led to believe there is no way to get it to work for now.

I have to ask, are you a moderator or someone actually involved in the app development. I am hoping to speak to Andy, who I know could easily confirm or deny the ability of the screen saver module at this point. I know that the Screen Saver is Supposed to support audio input. The documentation says so. If it is broken, the best thing would be for Andy to know.

[juxtiphi]

with all the ways you have looked at it i am led to believe there is no way to get it to work for now.

I have to ask, are you a moderator or someone actually involved in the app development. I am hoping to speak to Andy, who I know could easily confirm or deny the ability of the screen saver module at this point. I know that the Screen Saver is Supposed to support audio input. The documentation says so. If it is broken, the best thing would be for Andy to know.

I am just a moderator.


but I did supply youwith a quote from key comton who is a site administrator and I believe he knows all about the app and its capabilities,

this is the quote

Unfortunately the Mac OS does not allow G-Force to access the audio driver information -- however, if you have an iSight or any other microphone device, you will find that this method works very well.

take it for what its worth as I dont know what system you run on your laptop but if its mac then theres your answer.

[ciphex]

hehe, well i know that is an answer in theory man... I found that in the docs myself before ever posting here. The problem is it's just not working for me as the docs and that quote claim that it should.

so my mission is to figure out WHY it's not working when it should, make Andy aware of it so that
a) it can be fixed
b) I can do a blog write up on how to pimp your mac with Gforce awesomeness.

I was hoping to get some developer feedback by posting here but I am sure they stay busy doing the dirty work so possibly I will shoot someone an email directly unless you think the dev(s?) might run across this here sooner or later?

I do appreciate the dialog and your help.

also: you made reference to a thread in your first reply that discussed this topic. i would love to read that thread completely. the link you posted for it was broken, could you please toss that out for me again?

Many thanks

[juxtiphi]

hehe, well i know that is an answer in theory man... I found that in the docs myself before ever posting here. The problem is it's just not working for me as the docs and that quote claim that it should.

so my mission is to figure out WHY it's not working when it should, make Andy aware of it so that
a) it can be fixed
b) I can do a blog write up on how to pimp your mac with Gforce awesomeness.

I was hoping to get some developer feedback by posting here but I am sure they stay busy doing the dirty work so possibly I will shoot someone an email directly unless you think the dev(s?) might run across this here sooner or later?

I do appreciate the dialog and your help.

also: you made reference to a thread in your first reply that discussed this topic. i would love to read that thread completely. the link you posted for it was broken, could you please toss that out for me again?

Many thanks

I dont think can help you.

please use the search function and search for the keyword screensaver look on the second or third page of results for a thread entitled screensaver

Key compton is not just someone who trys to help GF users as I do, he is involved with the development and if he says it cannot be done then I am inclined to believe him. and if he's aware of your request then Andy is also.

although it might be possible the docs are sketchy at best where the SS is concerned as I could not find one line that speaks to making the SS react to music.

and I still dont know why you wouldnt want to use the standalone for this.

sorry i dont have the technical knowledge to help you further.

[ciphex]

Well I have managed to get it working.

It seems that when running in a dual monitor setup and telling the screen saver to run on all monitors then two or more instances of gforce are launched. Since one instance already has the mic input reserved then errors are thrown. The only solution for now is to set the screensaver to run on only one monitor.

[juxtiphi]

Well I have managed to get it working.

It seems that when running in a dual monitor setup and telling the screen saver to run on all monitors then two or more instances of gforce are launched. Since one instance already has the mic input reserved then errors are thrown. The only solution for now is to set the screensaver to run on only one monitor.

Thats not how it works in XP, the option for showing the screensaver on 2 screens just splits GF down the middile, at least thats what it does on my screens, so only one instance of GF is running . And I have my video card set to show dual desktop, personally I didnt really like the way it looked split down the middle like that but I dont use a screen saver so its no biggie for me. I'm to busy at my comp most of the time for it to be worth setting up the SS although I now see how it would be cool.


who knows maybe Andy will implement a few more settings for the screensaver in future versions.

[ciphex]

yeh i don't think i would like it split down the middle either.

I decided to combine the things i learned working thru this into a guide for people wanting to do the same sort of thing:

http://www.soundspectrum-forums.com/vie ... php?t=6091