Hmmm, I notice it too...I THINK that if you go the savestate menu (then make a 
save),
the sound issue appears. HOWEVER, I think it goes away if you go into the 
savestate
menu once more (don't make a save) and then back out, but I could be wrong. But 
I'm
pretty sure this issue may be harder to fix than I think it is.

[deleted comment]

There is ALSO a slight buzz noise that occurs when going from game to sub-menu. 

Just thought I would mention it, hope that helps!

Here's something very interesting I recently noticed about the emualtor's sound
issues.....Currently, both Snes9xGX and FceuGC both use similar menu code, 
right?
Now, while they are two different emulators, they both open a menu when you 
press
"Z". On Snes9xGX, when you go to a game, then to the menu, and back to the 
game, the
sound seems to get tinny/raspy (as if it's downsampled to 22000Hz instead of the
normal 32000Hz). However, in FceuGC, when you do the same thing (game to menu 
and
back to game) the sound isn't affected at all. Shouldn't this be the case in 
Snes9xGX
as well? Or is this an issue the HBC developers have to deal with? Or is this an
issues in the Snes9xGX sound code itself? I'm really confused about the whole 
thing....

Anyway to pinpoint this issue?

Or rather, is this an issues with libogc as a whole and simply cannot be fixed?

is this still happening on the latest version? if it's not happening on vba or 
fceu 
it could have something to do with switching between 32khz and 48khz

[deleted comment]

Yeah, it's still happening, so, it could be that one, the .ogg that plays in 
the menu
is 48KHz and two, the Super NES is at 32KHz? And as a result of going back and 
forth,
the sound gets all messed up?

I also had to switch back and fourth 3 to 4 times from the menu to compare the 
sound. 

Can anyone else confirm this?

Do we have any updates? 

Still doing it with 4.1.2. Kinda annoying but not a super big deal... just have 
to
enter/exit a few times.

The next version of Snes9x wii (I think) will have a new sound core (Blaarg's 
new
core, if I'm not mistaken...*sigh* I can never spell his name right).  

If I downloaded the source code, the .ogg file sample frequency could be 
changed to
32000Hz instead of 48000Hz. Is that the frequency it's set at? The SuperNES 
sampling
rate is fixed 32KHz and when you go to the menu, it switches to 48KHz (either 
the
game menu or setting menu, the sample rate is 48KHz, even if there's no music 
being
played).  This way, the tinny sound would definitely go away if the change 
could be
made.  Unfortunately, either my ISP is being retarded or some other setting is
preventing me from downloading any of the SVN revisions. At least one of the two
Snes9xWii sound issues could be resolved.  If for some reason I am unable to do 
this,
is there anyone out there that want to put it to the test? 

[deleted comment]

Not quite that simple. the sound effects would also need to be changed to 
32khz, and 
libasnd is hard-coded to 48khz. And this is all assuming this IS the problem.

[deleted comment]

Now that Snes9x 1.52 is released, at least the synchronization will be improved 
on
Snes9xGx....

I should post this here too:

It's a libogc issue, with the audio init code.
the issue is when changing the sampling frequency from 48Khz to 32Khz - libogc 
just 
flips the bit, the SDK does a whole bunch of stuff

power your Wii off for at least a few seconds. then try running HBC via bootmii-
boot2, then "return to system menu". you'll see the same issue.

I know it's a libogc issue, but isn't there ANYthing that can be done about 
that? 

[deleted comment]

shagkur has graciously fixed this in libogc for us. please test this beta, I 
believe 
it is now fixed. Let me know.
http://tantric.dyndns.org/snes9xgx-fix.zip

[deleted comment]

Your links dead, by the way.

sorry here it is again: http://tantric.dyndns.org/snes9xgx-beta5.zip

[deleted comment]

I tested Snes9x 4.1.8 (beta 5) for approximately 90 minutes and the sound 
frequency
issue is still occurring (when switching back and forth between the menu and the
game). I'm not doubting that the issue was fixed in libogc, but is there 
something
that I'm not doing right?  

I confirm this issue still present in 4.1.8

Yes, and if you go back and look at the SVN, you can clearly see that the issue 
was
reported as being "fixed" in libogc.  Either I didn't read this correctly, or I 
can
be considered a skeptic. I already had to make another issue trying to get 
dborth's
attention.  Seems as though posting THAT was in vain. :-|

getting my attention won't help get it fixed faster. Nothing I can do to fix 
this, 
and it's a libogc issue I'm sure.

So, what was this about? "shagkur has graciously fixed this in libogc for us." 
I'm
confused...

I thought it was fixed, shagkur hoped it was too, if it isn't, it isn't. sorry.

It's okay. Is there a reason why it's noticeable in Snes9x, but not in FceuGC 
and VbaGx?

probably because snes uses 32khz sound, which is what the actual SNES had too. 
Convenient that the Wii also supports 32khz!

Then that would mean FCEUltraGC and VbaGX use a sampling frequency of 44100Hz, 
right?
 If I knew how complex and frustrating this issue would end up becoming, I might have
hesitated bringing it up in the first place.....sorry. :( 

someone please try this patch: 
http://sourceforge.net/tracker/?func=detail&aid=3110188&group_id=114505&atid=668
553

Perhaps someone can try this and let me know if the tinny bug is still present?

http://tantric.dyndns.org/snes9xgx-beta.zip

Whoa, didn't even notice this beta release. Will give it a try ASAP. 

Hmmm.....Looks like it still happens, but maybe not as frequently....? I wish 
other people can test this. Oh, and EWJ2 runs beautifully. 

Has anyone been able to test this build out?

atleast part of this issue is asnd's lack of interpolation when resampling non 
32k/48k sample data

I don't think something like this can be fixed; I thought it was an issue with 
libogc in general.