383 replies to this topic

[ZXDunny]

Definitely not Wifi - I have mine permanently disabled.

D.

[Caine]

Definitely not Wifi - I have mine permanently disabled.

I can confirm this. At all times when I noticed this sound, WiFi was disabled.

[Ivanovic]

Nope it's definitely a boing, like a ruler being twanged on a desk.

I've had that ever since I got my Pandora! It's quite light and quiet, but it's quite a surprise when it happens out of nowhere. And for the record, it's coming out of the speakers.

IIRC ED once mentioned that this would be the audio chip itself going into some energy savings mode since it is now unused. So you should not hear those "boings" unless audio is turning itself into some low power mode. This is no hardware defect but just some strange general behaviour.

[EvilDragon]

The boing is the audio amp releasing the remaining power before it shuts itself off.
That's normal with every Pandora and definitely not the same as the annoying buzzing.

[Caine]

That's good to know.

[Ivanovic]

known broken programs/games:

• Picodrive - broken sound and video
• Snes9x4p - doesnt work, just quits back to desktop after selecting game

Okay, I know that Snes9x4p does run if you are using a game where you changed the mode to something not hardware accelerated eg running the old kernel. The problem here is that the emulator segfaults once hardwarescaling via Notaz libsdl is used / switched to. This might be a generic problem for all progs relying on hardware accelerated libsdl features.

Besides, if you are running Snes9x4p with a non hardware scaling mode you will have broken audio. With other words: The symptoms look identical/similar to Picodrive, where audio and video are borked.

[jottt]

I did that, both the GUI and /proc/pandora/cpu_opp_max tell me my max OPP is set to 5
Edit: Ok, 880MHz is the highest I can go, as far as I remember that's the highest clock I could go at OPP3. Is it possible my Pandora's ignoring the allowed OPP somehow?

Remove the battery and leave pandora off for 10 minutes then.

@Fahrstuhl & Notaz

I can reproduce that behaviour over here. Mine usually does 1.1 GHz without problems. It crashes now directly in the settings dialogue if at any time before the 3.2 kernel was booted. Even the reflash to Beta1.1 did not change that.
Notaz' Fix helps however, als long as I don't boot again into Kernel 3.2.


Another (small) Script-Issue: When you set the maximum allowed speed to X, then set the current speed to the maximum allowed speed and afterwards lower the maximum allowed beneath the current speed you can't change the current speed anymore. It just replies with a textbox.

[coverilla]

Am I the only one who has problems with right SD slot (not working) when running any OS from SD card? This happened after I flashed SuperZaxxon beta 1.1 on my nand. When I boot from nand both slots are working fine.

[MarioPandio]

^ I didn't experience loss of the right SD slot when running beta 1.1, or the new kernal via SD. However I seem to be the only person for whom flashing the new bootloaders disables the touchscreen on the nand installations.

[Linux-SWAT]

Seems the internal wifi of my non-wifi pandora works with the new kernel .
Will test further.

[milkshake]

Seems the internal wifi of my non-wifi pandora works with the new kernel .
Will test further.

lol

[notaz]

Am I the only one who has problems with right SD slot (not working) when running any OS from SD card? This happened after I flashed SuperZaxxon beta 1.1 on my nand. When I boot from nand both slots are working fine.

Yeah the new bootloaders set up some things for SD2 that pre-SuperZaxxon kernels can't handle. I'll edit my posts to add a warning.
You can upgrade your SD installation too to SuperZaxxon, it should be stable enough.

I didn't experience loss of the right SD slot when running beta 1.1, or the new kernal via SD. However I seem to be the only person for whom flashing the new bootloaders disables the touchscreen on the nand installations.

Sounds like a faulty board.

[Ian J]

The boing is the audio amp releasing the remaining power before it shuts itself off.
That's normal with every Pandora and definitely not the same as the annoying buzzing.

Wasn't really an issue more an observasion. But it's good to know.

Thanks ED

[Linux-SWAT]

So on 3.2 the signal strength is better than with 2.6.
Dmesg is clean,
No disconnections.
But pretty slow :/. I have no good AP to bench.

I rebooted on 2.6 just in case, indeed wifi doesn't work.

Dunno if already reported, but my usb network adapter doesn't work, dmesg stays silent.
ID 0b95:7720 ASIX Electronics Corp. AX88772

[milkshake]

strange that it fixed your wifi?

[Linux-SWAT]

It made it stable, it's certain.
It caused OS crashes on 2.6.
That's also why i was worry about blacklisting modules.

[McLovin]

so, the wifi driver has changed? didn't want to make a fuzz of it because i hadn't used my pandora for so long, but my wifi-experience has drastically improved since I'm using the new kernel.

[Linux-SWAT]

A lot more than this has changed since.

[notaz]

Well wifi driver was more or less backported to 2.6, and doesn't differ much..
It could be that OMAP power saving helps wifi module a bit (less interference perhaps?), and fixes for base OMAP code could help too. In my own tests wifi hasn't changed at all, still throttles itself to 70kB/s or so.

[Ivanovic]

known broken programs/games:

• Picodrive - broken sound and video
• Snes9x4p - doesnt work, just quits back to desktop after selecting game

Okay, I know that Snes9x4p does run if you are using a game where you changed the mode to something not hardware accelerated eg running the old kernel. The problem here is that the emulator segfaults once hardwarescaling via Notaz libsdl is used / switched to. This might be a generic problem for all progs relying on hardware accelerated libsdl features.

Besides, if you are running Snes9x4p with a non hardware scaling mode you will have broken audio. With other words: The symptoms look identical/similar to Picodrive, where audio and video are borked.

Okay, with the latest release (1.39ff.20120226.1) it will at least be possible to start the snes9x emulator with the experimental kernel. Audio is still broken and if you switch over to a hardware accelerated scaling mode it will crash. At least those players who don't want audio anyway should now be able to play using the software scalers and the experimental kernel...