Fatdog64-802/801/800 Final [21 May 2019]

[vtpup]

Relevant 2 sections of var/log/messages:

Mar 31 02:19:53 GFD4627 user.info kernel: input: PC Speaker as /devices/platform/pcspkr/input/input5
Mar 31 02:19:53 GFD4627 user.notice kernel: cfg80211: Loaded X.509 cert 'sforshee: 00b28ddf47aef9cea7'
Mar 31 02:19:53 GFD4627 user.info kernel: Intel(R) Wireless WiFi driver for Linux
Mar 31 02:19:53 GFD4627 user.info kernel: Copyright(c) 2003- 2015 Intel Corporation
Mar 31 02:19:53 GFD4627 user.err kernel: Error: Driver 'pcspkr' is already registered, aborting...

...


31 02:19:53 GFD4627 user.info kernel: input: chtmax98090 Headset Jack as /devices/pci0000:00/808622A8:00/cht-bsw-max98090/sound/card1/input8
Mar 31 02:19:53 GFD4627 user.info kernel: intel-spi intel-spi: w25q64dw (8192 Kbytes)
Mar 31 02:19:53 GFD4627 user.notice kernel: Creating 1 MTD partitions on "intel-spi":
Mar 31 02:19:53 GFD4627 user.notice kernel: 0x000000000000-0x000000800000 : "BIOS"
Mar 31 02:19:53 GFD4627 user.err kernel: Error: Driver 'pcspkr' is already registered, aborting...
Mar 31 02:19:53 GFD4627 user.info kernel: Console: switching to colour frame buffer device 170x48
Mar 31 02:19:53 GFD4627 user.info kernel: i915 0000:00:02.0: fb0: inteldrmfb frame buffer device
Mar 31 02:19:53 GFD4627 user.info kernel: input: HDA Intel PCH HDMI/DP,pcm=3 as /devices/pci0000:00/0000:00:1b.0/sound/card0/input9
Mar 31 02:19:53 GFD4627 user.info kernel: input: HDA Intel PCH HDMI/DP,pcm=7 as /devices/pci0000:00/0000:00:1b.0/sound/card0/input10
Mar 31 02:19:53 GFD4627 user.info kernel: input: HDA Intel PCH HDMI/DP,pcm=8 as /devices/pci0000:00/0000:00:1b.0/sound/card0/input11

Available soundcards -- screenshot.

[kirk]

vtpup,

My only suggestion is to try each of: card 1 device 0, card 1 device 1 from Set Default Sound card. After choosing one open AlsaMixer and make sure that it's not muted. Muted controls wil have MM at the bottom. Press the m key to toggle mute/un-mute. I've attached a picture of AlsaMixer that has the Master contol muted, just in case you haven't had to deal with AlsaMixer before.

[jd7654]

This is interesting: I just fried my laptop speaker with ALSA!

Didn't think that could happen. Was just adjusting some levels in alsamixer, and I heard some beeps and pops. And then the smell...that unmistakable burning electronics smell. So I lifted up to see what was going and where the smell came from. And felt the area below speaker was too hot to touch, could have burned my finger if I left it there.

Now one speaker is fried.

OK, that's enough playing with Linux sound on this R11. Now I gotta open this up and repair/replace the speakers. Hopefully no other damage.

Cautionary tale...

[stemsee]

This is interesting: I just fried my laptop speaker with ALSA!

Didn't think that could happen. Was just adjusting some levels in alsamixer, and I heard some beeps and pops. And then the smell...that unmistakable burning electronics smell. So I lifted up to see what was going and where the smell came from. And felt the area below speaker was too hot to touch, could have burned my finger if I left it there.

Now one speaker is fried.

OK, that's enough playing with Linux sound on this R11. Now I gotta open this up and repair/replace the speakers. Hopefully no other damage.

Cautionary tale...

I reported this happening to me on DebianDog about 2 months ago. I realised that it happened after boosting a gain signal in alsa for a soc sound card btrcr_rt5640. Both speakers and the screen edging melted, could have burst into flames and burnt the house down ... no shit!

[jd7654]

I reported this happening to me on DebianDog about 2 months ago. I realised that it happened after boosting a gain signal in alsa for a soc sound card btrcr_rt5640. Both speakers and the screen edging melted, could have burst into flames and burnt the house down ... no shit!

Yeah, this is insane! I thought the same thing: if this was on a tablecloth and I didn't catch it, would have caught fire. Wow...imagine that: you could hack into someone's laptop, tweek the speaker controls, and burn their house down! "death by stereo."(1987)

Why does the audio speaker circuit have that much juice, available via a simple software command? This one is a chtmax98090 chip in the Acer R11.

[jake29]

I reported this happening to me on DebianDog about 2 months ago. I realised that it happened after boosting a gain signal in alsa for a soc sound card btrcr_rt5640. Both speakers and the screen edging melted, could have burst into flames and burnt the house down ... no shit!

Yeah, this is insane! I thought the same thing: if this was on a tablecloth and I didn't catch it, would have caught fire. Wow...imagine that: you could hack into someone's laptop, tweek the speaker controls, and burn their house down! "death by stereo."(1987)

Why does the audio speaker circuit have that much juice, available via a simple software command? This one is a chtmax98090 chip in the Acer R11.

I noticed when running Ubuntu on my laptop, max SoC audio volume is quite low - perhaps there is a known danger around pushing the chip too hard.

[jd7654]

I noticed when running Ubuntu on my laptop, max SoC audio volume is quite low - perhaps there is a known danger around pushing the chip too hard.

Not sure if that's related or not.
Normally, if your audio chipset is supported, getting your laptop speakers working is just a matter of selecting the right card, usually the one that says analog. And then adjusting alsamixer (setting levels, unmuting) and maybe map the master volume slider to the tray, etc. I don't think simply adjusting volume would cause your laptop to melt.

One final note of warning:
The standard alsamixer picture above that kirk posted is what I normally see on other laptops. But if your alsamixer looks like the picture below, individual controls/pages, over 100 of them, beware. I was playing around with some setting over past page 100 or so, like the last snapshot, some kind of gain, and that's when it began to smoke. Can't remember exactly which one, because I shut off when it began to smoke. Then when I booted up and it started getting hot again, I shut it back off. So I realized what was happening and deleted the fatdog savefile and the overheating went away, along with one speaker.

[vtpup]

Hmmm.....there's a report of similar speaker heating problems here:

https://github.com/GalliumOS/galliumos- ... issues/270

worrisome....

[jd7654]

Hmmm.....there's a report of similar speaker heating problems here:

https://github.com/GalliumOS/galliumos- ... issues/270

worrisome....

Yeah, scary. Looks like a few of them got hit with blown speakers while working on GalliumOS.

I never heard of this issue till I fried my own speaker. Wish I had known about it or it was more widely disseminated. Googling around, there were reports of this here and there but didn't seem to garner much attention.(at least nothing like Galaxy Note 7 level)
I like this one, semi joking, "How To" from many years ago:
How to fry speakers in your Chromebook

Tinkering is fun and all that...until you burn through your desk!

[jamesbond]

I'm speechless.

Even ignoring all other aspects of the kernel/hardware bug, doesn't those who designed the codec chip think of over-current protection? It's just an extra transistor and a resistor ... something that virtually cost nothing in an IC.

Another thing that made me speechless is this: the link you gave indicated that the bug existed since Dec 2012, if not before. And nobody bothered to fix it? Not HP, not Acer, not Google, not anyone?

At this point in time then the only sure way to avoid the bug is run Chromebook without sound support

Anyway, thank you for your warning and sharing your experience. It's bad what happened to your Chromebook, I hope you can get it fixed one way or another.

I'd probably re-posted this in my blog to serve as warning for others. I came from the days where re-programming the CRTC the wrong way can physically destruct the CRT monitor, but like you I believed that those days are long past. Apparently it isn't

[rcrsn51]

At this point in time then the only sure way to avoid the bug is run Chromebook without sound support

See my suggestion on page 4.

[jd7654]

Another thing that made me speechless is this: the link you gave indicated that the bug existed since Dec 2012, if not before. And nobody bothered to fix it? Not HP, not Acer, not Google, not anyone?

Yeah, I'm surprised in this day and age that nobody lawyered up or went to the press. Guess there was no financial incentive. I wonder though, if it is just more of packing more power into ever smaller size, no margin for error.(like batteries)

Anyway, thank you for your warning and sharing your experience. It's bad what happened to your Chromebook, I hope you can get it fixed one way or another.

Thanks. Could've been worse. Stemsee seems to have been hit harder, with both speakers and screen edge melted. At least I still got one working, but I'm surprised the plastic didn't melt, felt like a flameless lighter was underneath the grill!

I took a look at the iFix it for the R11 Speaker Replacement and it doesn't look too hard to replace the speakers...which are on either side of the lithium ion battery! <omg>

[don570]

rcrscn51 recommended external sound card

I bought a cheap sound card adaptor ($10) for my raspberry pi
and the quality of sound was excellent and easy to set up with alsa mixer.

http://murga-linux.com/puppy/viewtopic. ... 602#998602

[Sage]

... external sound card
I bought a cheap sound card adaptor...

Why compromise?! A small solid state hi-fi amplifier can be had from China/eBay for ~same cost. Plug into green sound OP, just need a pair of good speakers, then. But if you can't afford expensive ones pick a pair of old ones out of your local amenity dump/ skip/dumpsters. Get as large as possible in wood cabinets. If they're cheap infinite baffle types [ie totally sealed] take them apart, reinforce them with scrap timber strips (wood glue will suffice}. Stuff the internal space with scrap polystyrene foam [used for packaging of eg white goods, etc]. Firmly screw and stick the whole lot back together, possibly using foam draught excluder along the joints. Bingo - might even have a few pennies change from your £10/$10/E10.
Always a better choice than the internal /commercial options.
If you're really strapped for cash, it is possible to convert an old ISA sound card into a suitable amplifier with a bit of circuit tracing and modification, powered from a USB port. I described the details several centuries ago. But, the master stroke is always good speakers.

[kirk]

Even ignoring all other aspects of the kernel/hardware bug, doesn't those who designed the codec chip think of over-current protection? It's just an extra transistor and a resistor ... something that virtually cost nothing in an IC.

Hard to believe there no current limiter, just wow.

[vtpup]

My guess is that the move toward general purpose programmable input-output chips has led to sloppy design. Basically stick some multiprocessor in there and tell it what to do and how to configure itself via easily altered firmware and you cut manufacturing costs for dedicated single purpose devices. But the very programmability of firmware is its danger. It can be anything, and do anything. Yes you could put a current limiter onboard, but if your mindset is hey this is hard programmed not to exceed current specs I don't need to worry about that, nope. Everything is mutable. Firmware updates routine. Goof your update code or delivery and bad things can happen. The internet of things will bring more of these problems.

As far as the specifics for the Acer R11, it's fine to add an external card and/or speakers for a desktop system, but for a highly portable system like a Chromebook, that's impractical.

I do wonder if those of us able to open the case could add our own current limiter to the speakers. A patch, I know, but I'm still interested in getting this thing useful.

[don570]

For people writing software for fatdog64 here is a tip...

A simple warning message with yaf-splash to show on screen until clicked. Note that quotes have to be protected.

Code: Select all

yaf-splash -timeout 0   -placement center  -bg orange  -bw 2 -outline 0 -margin 15  -text "EXAMPLES

In the following examples I'll be splitting One Click.mp3 into folder named split.
    To split between specific time range...

In this case just add the time points where you wanna split. 
e.g. 15.00 30.00 would produce a 15 min track extracted from original file beginning at 15 min and ending at 30 min.
mp3splt \"One Click.mp3\" -d split 15.00.00 30.00.00


Here's an explanation of the recommended options:
    -f: for MP3 files only, increases precision and is needed if the MP3 files are variable bit rate (VBR).
    -t TIME: specifies the length, measured in time, to make each piece. 

You will replace `TIME` with a numerical value expressed in minutes, such as 4.0 for four minutes or 7.30 for seven minutes, thirty seconds. 
In our example, we picked four minute pieces, so the command line will be
mp3splt -f -t 4.0 -a -d split *.mp3

 " 

[don570]

I noticed something about sound in fatdog...

There is a difference in how BarryK Easy pyro linux boots ALSA configuration...

Here is the configuration in /etc/init.d/10-alsa
https://github.com/puppylinux-woof-CE/w ... 10alsa#L13

amixer -s -q <<EOF
sset Master 75% unmute
sset Master -12dB
sset 'Master Mono' 75% unmute
sset 'Master Mono' -12dB
sset 'Master Front' 100% unmute
sset 'Master Front' -12dB
sset Front,0 75% unmute
sset Front,0 -12dB
sset PCM 90% unmute
sset PCM 0dB
sset Synth 90% unmute
sset Synth 0dB
sset CD 90% unmute
sset CD 0dB
sset Mic 0% mute
sset PCM,1 90% unmute
sset PCM,1 0dB
sset Wave 100% unmute
sset Music 100% unmute
sset AC97 100% unmute
sset 'Master Digital' 75% unmute
sset DAC 90% unmute
sset DAC -12dB
sset DAC,0 90% unmute
sset DAC,0 -12dB
sset DAC,1 90% unmute
sset DAC,1 -12dB
sset Headphone 75% unmute
sset Headphone -12dB
sset Playback 100% unmute
sset "SB Live Analog/Digital Output Jack" off
sset "Audigy Analog/Digital Output Jack" off
sset Speaker 75% unmute
EOF

Compare to /etc/init.d/99-alsa in fatdog64

_____________________________________________________________

amixer -s -q <<EOF
set Master 75% unmute
set Master -12dB
set Speaker 100% unmute
set Speaker 0dB
set Speaker+LO 0dB unmute
set 'Dock Mic' 0db unmute
set 'Headset Mic' 0db unmute
set 'Master Mono' 75% unmute
set 'Internal Mic Boost' 20db unmute
set 'Master Mono' -12dB
set Front 100% unmute
set Front 0dB
set PCM 100% unmute
set PCM 0dB
set Synth 100% unmute
set Synth 0dB
set CD 100% unmute
set CD 0dB
set Mic 0% mute
set PCM,1 100% unmute
set PCM,1 0dB
set Wave 100% unmute
set Music 100% unmute
set AC97 100% unmute
set 'Master Digital' 75% unmute
set DAC 90% unmute
set DAC -2dB
set DAC,0 90% unmute
set DAC,0 -2dB
set DAC,1 90% unmute
set DAC,1 -2dB
set Headphone 100% unmute
set Headphone 0dB
set Playback 100% unmute
set HDMI 100% unmute
set HDMI 0dB
set S/PDIF 100% unmute
set S/PDIF 0dB
set "SB Live Analog/Digital Output Jack" off
set "Audigy Analog/Digital Output Jack" off
EOF

Maybe speaker and headphone are set too high -->

Code: Select all

  set Speaker 100% unmute
set Speaker 0dB
set Headphone 100% unmute
set Headphone 0dB

[vtpup]

Thanks Don570.

re. the overheated speaker issue -- I'm wondering if a simple safety workaround in my case might be putting a fuse inline with each speaker.

Any thoughts on how to size it?

[Sage]

putting a fuse inline with each speaker.

Any thoughts on how to size it?

In the olden days, quality hi-fi sometimes had a panel on the rear of the amplifier with small fuses for the speakers. Think they were ~ 100s of mA but will check later. Notwithstanding, the sizing of fuses has always been a black art, with little agreement and conflicting 'calculations' between advocates. There's a load of theory about exploding wires, but whether it's a true science in practice is debatable.