I pretty much succeeded in getting things to work, the wireless LAN required the intel drivers, which were not included in the install at that time, but downloading and compiling the drivers went smoothly, so in no time I had wireless networking working with knetworkmanager supporting it, and connecting to open WLAN’s WEP protected WLAN’s, and my home WLAN which runs WPA.

Bluetooth also worked, and I tried connecting to my Mobile phone (That’s Cell phone for americans), and downloaded a couple of pictures I had taken with it. The mobile phone is a Nokia 6310i by the way.

But no sound.

And even though I found lots of discussion,especially on ubuntu fora, about sound with hdaudio from intel, I was not able to get it working even though I upgraded to the latest version of ALSA.

Very recently though I found a forum, where a guy with the signature “oldcpu” guided someone else through a step by step test and config of the sound configuration, and following his suggestions I soon had sound working.

So here is my setup (I gave up getting yast to configure sound for me, and manually edited the file instead):

This is the content of my /etc/modprobe.d/sound:

options snd-hda-intel model=toshiba
alias snd-card-0 snd-hda-intel
alias sound-slot-0 snd-hda-intel
I haven’t experimented with it at all since it started working, but I think the “toshiba” option was the trick for me.

Other than that, I have ALSA 1.0.15 driver and tools, manually installed and compiled. But since then yast has upgraded the driver to ALSA 1.0.15 as well, and it still works. But yast breaks sound for me, and so does alsaconfig.

A very good test for me while testing the options in the file above was this (also suggested by “oldcpu”):

speaker-test -c2 ddefault -twav

Suddenly the speakers were saying: “front left”, “front right”, and so forth endlessly until i killed it with ctrl-c

So If anyone else is having trouble with this, I hope the above gave you some inspiration about how to get sound support working on your laptop.

Der var gløgg og pebernødder

And indeed Leif Lodahl has submitted a link to a manual for Danish authorities explaining how to handle the upcoming trial where they must be able to read both ODF and OOXML documents starting 1. January 2008.

In that manual I found the following section:

1.3.3 Hvad betyder det, at standarden er åben?
Standarden skal være:
> fuldstændigt dokumenteret og offentligt tilgængelig,
> frit implementérbar uden økonomiske, politiske eller juridiske begrænsninger på
implementering og anvendelse, hverken nu eller i fremtiden, og
> standardiseret og vedligeholdt i et åbent forum via en åben proces
(standardiseringsorganisation).

Translation:

1.3.3 What does it mean, that the standard is open?

The standard must be:

> Completely documented, and publicly available,

> freely implementable without economical, political or legal limitations on

implementation and use, whether now or in the future, and

> standardized and maintained in an open forum via an open process

(standardization organization)

From this (and having read some of the technical comments from the Danish national board of the ISO standardization committee which rejected the fast track approval of MSOOXML a few months ago) I think it is pretty clear that MSOOXML does not meet the criteria for an open standard as described in that manual.

That would make the manual misleading!

Let me point out what I mean:

quote: > Completely documented, and publicly available,

Comment: MSOOXML refers to closed MS documents, and therefore fails on both the “completely documented” and “publicly available” points.

quote: “

> freely implementable without economical, political or legal limitations on

implementation and use, whether now or in the future, and“

Comment: Several projects have been trying, or are trying to write plugins or converters for MSOOXML to ODF, but as far as I know none have so far succeeded. Partly because there are things in OOXML which can not be mapped to ODF. Partly because it is a huge standard and probably never intended to allow true interoperability.

Further, there are still patents on parts of the MSOOXML specification, and therefore it fails on the “freely implementable without economical, political or legal limitations” point.

quote: “> standardized and maintained in an open forum via an open process

(standardization organization)“

Comment: MSOOXML also fails on this last point, at least it is not currently being maintained in an open forum. Perhaps if Microsoft were forced (for instance if every major office user on the planet decided to mandate truly open standards as the Dutch government and others have recently done) they would comply with the requirements for an open standard. But realistically, I think hell will freeze over before Microsoft does this.

As I wrote in a lengthy earlier post, I am pretty much happy about my mythtv setup now. It has been running quite well for some time now.

But one thing has been bothering me: I had not managed to configure the box to turn itself off, and back  on, so I let it run 24/ 7, which ofcourse is fine from a purely stability point of view, after all this is linux which unlike windows is fine to run for years without reboot.

But from a power consumption point of view, this has been annoying me and today I set out to fix it.

So today I set out to find out why I could not get mythtv to do this right. I remember that last time I looked at this (which is a long time ago) I tested acpi wakeup from a term, and was encouraged to find that it worked. But it did not work when mythtv shut down the box, and I did not have time to pursue the problem back then.

Until now.

I found out that my script to set the ACPI alarm (by writing to the virtual file /proc/ACPI/alarm) was not getting called correctly from mythbackend.

The error was actually quite trivial, as I had mythtv call my script with $time, in time_t format and stated like this: $($TIME) + 300).

First I saw on the terminal as the “shutting down” messages were whizzing by, that this was literally sent as: “2007/10/10 09:27 +300″ which made the script which was supposed to convert the time in local time format to UTC before writing it to /proc/acpi/alarm.

I then changed the format in mythtv from time_t (I am not sure how to get the 300 seconds extra bootup time originally intended) to YYYY/MM/DD HH:MM, and now the box shuts down and boots up as needed to record the shows I have scheduled.

I think I took my script (with trivial modifications) from an example on this page:  here but they have a different script now. I haven’t tried the new script, but it looks ok as far as I can see, and will work with “old kernels” like the 2.6.18-5 I am using for this box, as well as the 2.6.22+ kernels which have changed the way acpi wakeup works. Do try it out, and let me know if that works for you. I don’t know when I will upgrade my mythtv box, as with this little annoyance fixed, I am actually quite happy with my mythtv box.

But one of these days, I have to find out why the remote does not work (it did once, but I reinstalled and upgraded from 10.1 to 10.2 since then, and probably changed some configuration somewhere, but that is a task for another day).

Some time ago I wrote that the PSU had unexpectedly died. A few days ago, I finally received a replacement PSU free of charge (apart from the 50 dkr it cost me to send in the defect one of course).

I had already found out that wlan now works with my Allnet 0271 nic (using wep only, since I can’t get it to connect without using knetworkmanager, and knetworkmanager refuses to let me select wpa. I finally found out that the prism54 driver which I use for the nic has incomplete wpa support, so I settled for wep for now).

I wanted to see if I could get the Hauppauge win tv nova-t dvb-t tuner to work, since everywhere it is stated that it should just work out of the box with opensuse 10.2. Well not quite, but after a long time trying to select some cx2388-something drivers with yast, and sometimes manually writing these drivers into /etc/modprobe.d/tv, and back and forth trying out different combinations, I finally tried to let kaffeine see if it could find any channels. And to my surprise it did !!!

It turns out, that kaffeine identified the tuner and on it’s own decided to use the cx22702 driver to control it.

So kaffeine successfully found the DVB-T multiplex in Denmark even when yast channel tune, and mythtv channel tuner could not. I am very grateful right now that I had kaffeine installed.

After this, I went back into yast, and selected the cx22702 driver instead of the default bcom-something (I turned off the other machine because it is getting late, and only on second thought decided to write about my partial success before I went to sleep. An update will follow sometime later, probably next week).

And then also mythtv could successfully tune and find the multiplex. (I had to tell mythtv to search using the german channel range, there is no danish list selectable in mythtv).

The parts I did not get to work tonight was audio, and perhaps IR (I say perhaps, because I will check if the batteries are alive before I draw any conclusions about that first

But I am pretty sure that will come along too soon enough.

Also I experimented a little with the nvidia-setting (nvidia xserver control), to set up the tv out so the text would be readable on the tv. I did not find a setting to my liking before I quit.

But theoretically I could start recording shows starting tomorrow, so I am optimistic that the remaining minor problems will be solvable too,

Update (22.07.2007 at 1:20)

Some time has passed, and I had resigned to use kaffeine to record and watch tv, since I could not find out how to get the correct drivers loaded automatically (I did so using modprobe instead), and mythtvbackend would crash every few minutes. If i had started mythtvbackend from a term, as root I saw this line when it crashed (I think it was the same line each time, but can’t be sure):

pcHDTV::GetSignal(fd 41 …….) (Invalid Argument)

I could not remember the parameters in the parenthesis so the “…” should not be taken literally

But tonight I played a little with it, and after having tried unsuccessfully to edit /etc/modprobe.d/tv manually to force Yast2 to accept the cx88_dvb and cx22702 modules to be loaded for the card (my Hauppauge winvt nova-t DVB-t), I found out that:

1. cx88_dvb depends on cx22702 and when cx88_dvb is loaded, then cx22702 is automatically loaded too.
2. I _could_ select cx88_dvb within Yast if I chose that _instead_ of the cx8800 module.

Now it seems mythtvbackend does not crash nearly so often as it did before. I guess the drivers I loaded before made mythtv think the card was a HD card?… I don’t know. I downloaded the source for mythtv with the intent to try and compile it, and see if I could make some sort of workaround, or at least provide better feedback to the maintainers if I filed a bug report. Now I will hold the bug report back a little until I am sure there is one.

I “fixed” one more small thing tonight:

Whenever thebox booted, it would halt with a message about the wrong screen mode being passed at boot, and I should choose another, or press space to continue anyway.

This is not the prettiest way to fix this, but simply removing the screen option, which was 0x31a, from grub (I did it from within Yast2->BootLoader) made it go away and just boot.

The downside of this simple and ugly fix, is that I do not have the flash screen active while booting, but at least it will now boot unattended,

I should probably fix this before too long though, as it most certainly will affect the WAF (Wife Acceptancy Factor;-) negatively if I don’t.

Still missing:

Configure mythtv to use the remote for controlling it.

There is one open document format, which is ISO certified, and that format is called ODF.

But since then Microsoft has managed to push it’s format which it calls open, and which name is OOXML through ECMA and now they are trying to fast forward this monster (6000 pages) of a format through ISO as well. This latest development regarding ISO has apparently been decided by some individual who felt like pushing it forward in spite of numerous objections to the standard.

A lot has been said about this since. But I think it is sad to read that it seems like it is not unlikely that the decision about which document format will be chosen in Denmark, will be not one open document format, but two formats, one open and one pretending to be open.

Needless to say (I should think, but I will anyway) the truly open format is ODF, and the one pretending to be is OOXML.

Apart from the fact that Microsoft’s biggest cash cow is their office suite, and that their OOXML refers to previous versions of the office suite from Microsoft, which invalidates the format from being called open, If Microsoft really wanted to “just interoperate and offer choice” they could just change the default file format of their office suite to ODF.
But of course Microsoft will not do this, because they are not interested in losing their cash cow. And they would, maybe not all at once but gradually if anyone could choose completely freely which program they wished to use to read and write documents and share those documents with others.

So I cross my fingers that ODF is chosen as the only standard to be used by the Danish state from january 1st 2008, but regretfully I have my doubts. It seems too many people here in Denmark are way too impressed with Microsoft (or afraid of them) to choose something Microsoft does not like us to choose.

Update (26.06.2007): i just read today that the IT spokes people of the various parties in the danish parliament have agreed to run a trial of one year starting january 1st. 2008, with both ODF and OOXML as supported document formats. Then in the first half of 2009 the trial must be evaluated.

I can’t believe how stupid that sounds to me. Either they don’t get that OOXML is _not_ an _open_ document format, and therefore does not qualify the requirements to even be considered, or they simply don’t care or dare not take the right decision.

So because of their incompetence or lack of courage or both the danish tax payers must now pay the price for postponing the decision for another year and a half. Because of this Microsoft have won no matter what happens in 2009. At the very least they will have been given 18 months in which they don’t have to worry about the danish state mandating open document formats, and in the meantime they can lobby the government until everybody thinks OOXML is really open and ODF is really not open. I know that sounds absurd, but so does this so called decision.

I am so disappointed about our government now.

Update (28.06.2007):

Just found this website today. I wish the IT spokes people mentioned above had read this 8and answered the questions truthfully, and drawn the necessary conclusions of course) before they agreed to use both ODF and OOXML for a year as a trial period.

Update (24.08.2007):

Member countries who are voting no to OOXML.

I love this report: Brazil has apparently had the guts to vote no, and not be lobbied or pressured by microsoft to approve of their defective “open” standard.

Poland apparently has also in one committee voted no, but for some reason (why do I fear foul play from Microsoft has something to do with it ?) another committee will be the one to make the decision on how Poland will vote.

Sadly germany has decided to vote yes. As for Denmark I sadly expect theDanish government to vote yes since it has already rubber stamped OOXML by deciding to “test” OOXML alongside Open Document Format starting next year. I cross my fingers and hope I will be wrong though, but it seems unlikely right now.

28.08.2007

I just found out that the Danish committee on standards (DS) has come to a consensus on what they will publish as their response on 03.09.2007. Until then apparently it is a secret.

I have checked from the article who the members are, and their possible relation to Microsoft:

CBS = Copenhagen Business School (I don’t know)

Microsoft = Microsoft…

Netcompany = (Microsoft Partner) , source:

Conzern = (Microsoft Partner), same source as for Netcompany.

Thydata = Microsoft gold certified partner, Source: here.

ITEK = Not Microsoft Partner.

IBM = Not Microsoft partner

KMD = Microsoft gold certified partner, source: here.

IT og Telestyrelsen (unknown relation to Microsoft)

NNIT = Microsoft partner, source: here.

TEctura = Microsoft global partner, source: here.

Ok, just found this link where somebody else did the leg work I started out on. Maybe I will finish my list later, but for now it seems that out of 30 members, 17 are close associates with Microsoft. See the article here.

Update: 30.08.2007,

“I just read this article titled: “Why OOXML Is Not Ready for Prime Time” by Jon Bosak, which I find gives several very good reasons for voting no to OOXML.

/Lars

But evening matches have now started, so from next week I hope the entire 4 man crew (myself included) will be racing with the other boats. so far the weather has been really fine for most of the still very young season of 2007.

After having spent the winter mostly doing other stuff, like getting the boat ready for the new season I recently decided to do something about my mythtv box which I had almost but not quite running as I liked in the beginning of december last year.

This is what I mean by “not quite running as I liked” by the way:

1. I did not get the ALLNET 0271 wireless NIC to work with OpenSuse, I felt I made some progress, but never quite got it to work even though I could connect from a knoppix cd (actually knoppix V 5.01) booted on the same machine.
2. I did not get the remote control configured so I used a keyboard and mouse to program recordings, and play them.
   1. I did confirm with Yast that key events got received correctly by the driver so I was probably close there.
3. I was not really happy with the boot screen, or the resolution when booting and running anything other than the mythfrontend.
   1. I had to manually bypass a warning about graphics modes during boot.
4. I did not untill later after the DVB-T tuner had stopped working for me, get the software installed for modifying the BIOS wakeup time, so I while I actually used it to record tv I always had to manually turn the machine on before the recording should start. As mentioned I later got this to work, so I am confident I will get it to work again, and will of course describe it here when I get to that point.

The box would not boot, probably due to the fact that I had installed on the SATA drive, and having the bios reset and not getting the right boot order. I finally gave up on it and reinstalled open suse 10.2, so now I have relative fresh install to start from. This is already a while ago in which I have not even powered up the box, so I hope I haven’t forgotten something essential.

Because this time I will try to record what I have done here in my blog so I and others may benefit from it. The progress might be slow since I get scouled at by my wife if I spend too much time by the computer, and I also have the boat in the water now. Oh, and then there is this half marathon thing early next month whivh I am training for. link: http://avis-maraton.net/.

This is the hardware in the box:

Motherboard: ASUS A8N-SLI premium

CPU: Athlon 3200+

TV tuner: Hauppauge WinTV Nova-T DVB-T model 909

Wlan: ALLNet 0271

OS: Open suse 10.2

The last time I installed this box I referred a lot to these web sites:

• http://wilsonet.com/mythtv/ This site has the headline Fedora Myth(tv)ology, and even though I am using open suse 10.2, I have found much useful information there.
• http://www.acaciaclose.co.uk/28654/index.html, This is actually the 3rd link returned by google for the search terms mythtv suse, and I have used this site a lot too last time.
• Another great place for Open Suse 10.2 customization: http://www.softwareinreview.com/cms/content/view/60/

I did not however succeed last time in getting my wlan ALLNET 0271 to connect to my WAP. I did try to install knoppix V (whatever was the latest version around november / december 2006) and that successfully connected to my WAP, so there certainly is hope. I just couldn’t get it to work with Suse 10.1 which I was using at that time, nor with Open suse 10.2 which I tried with the last time. Hopefully this time I will manage to get it right. (or buy another wlan nic which “just works” with Open Suse 10.2, time will tell).

This time I will try the suggestions from http://www.mythtv.org/wiki/index.php/Opensuse_10.2.

First thing I did those weeks ago after reinstalling, was to add repositories to Yast (the Suse package manager):

I found most of these as well as a description on how to add them at this site: http://en.opensuse.org/Additional_YaST_Package_Repositories

Then I went to nvidia’s homepage to get the binary nvidia driver for the graphics card. I added the nvidia package repository to Yast and let Yast install the driver for me.

Then I installed mythtv. This is where I start today.

While I have been writing the previous text, Yast has installed all the suggested packages in http://www.mythtv.org/wiki/index.php/Opensuse_10.2.#IntroDuction

Uhoh the first suggested test, namely >dmesg |grep ivtv, did not produce any output because the ivtv module had not been installed yet. So I tried (as root) modprobe ivtv, but then I got this error:

FATAL Could not load /lib/modules/2.6.18.8-0.1-default/modules.dep: no such file or directory

find / -name modules.dep revealed the problem: modules.dep was located in /lib/modules/2.6.18.8-0.3-default/modules.dep

apparently a newer kernel version, which uname -a confirmed was the running kernel.

Ok, do I really need to reboot to get this right (I did upgrade the kernel with libzypp earlier tonight, and it stalled forever at 99% so I eventually killed it) ?

Ok, a reboot made sure it looked in the right place for modules.dep, and I could modprobe ivtv successfully.

I rmmod’ed it again, because I want to see what Yast can do about the card. According to the section on DVB-T the Hauppauge Nova-T which I am using should just work with OpenSuse 10.2.

I went to Yast -> Hardware -> TV Card and configured the tv card which was listed as the following:

Listing:

0 – TV CX23880/1/2/3 PCI Video and Audio Decoder

0 – DVB Nova-T DVB-T model 909

I configured the first device (Video and Audio Decoder) like this:

Vendor, had “Autodetected” card highlighed and the Card was listed so I did nothing here, clicked “next”.

On the next screen I clicked “connected to” and selected the onboard sound card, clicked “next”

On the next screen I selected “Use IRC” and clicked “test”, I connected the IRC receiver to the tuner, and clicked a few keys on the remote and saw the correct keys get listed on the screen.

That was it for the decoder, and I clicked “next” and selected the DVB-T card (by the way just to avoid misunderstanding I only have this one DVB-T card installed in the machine. Why it gets listed as two devices I don’t quite know.

This card also had the vendor “autodetected correctly” so next screen:

Selected Audio “not connected”, next screen:

Hmm now I don’t see the screen from last time where I selected which drivers to use! ???

It’s late so I stop now.

Sunday 06.05.2007 (danish time notation

Today I found and installed nvram-wakeup, which at least on my motherboard (last time I tried this) is necessary to have the wakeup time set programmatically. Even though I think I sent my nvram config to the project the last time I dud this, I found that nvram could not identify my motherboard, and I couldn’t quite remember what I did last time to resolve this. But google helped me find this page: http://www.linuxtv.org/vdrwiki/index.php/Nvram_wakeup, where they suggest using the helper program “guess-helper” if nvram-wakeup can not detect one’s motherboard correctly. I did that the last time, and will now do it again. This will cost a number of reboots, with different manually set wakeup time in the BIOS each time. As the web site states, and I will restate: “Do this on your own responsibility” If you write bad values in your flash, you might need a hardware flash burner to revive your PC. You have been warned.

Update: (08.05.2007)

After I wrote the last entry I successfully (at least I think so, as I went through the BIOS detection procedure prompted by guess-helper, but did not actually test afterwards that it worked) installed nvram-wakeup, and configured mythtv to use it.

That involved instructing mythtv to use the time format: time_t, and setting the wakeup command to something like (I am writing from memory so the syntax might have slipped slightly, I will check this later…): nvram-wakeup -A -C /etc/nvram-wakeup.conf -s$(($time + 300))
I Used the command mythtvsetup as user mythtv to configure mythtv.

Then I configured mythtv to use the DVB tuner (several options were available, but I chose one which was mysetiously listed as “analog video” but which listed the name of the card as Hauppauge wintv nova-t DVB-T, so I thought I would give it a shot anyway as it was the best match apart from the “analog” part…

I shut down the machine, disconnected it and carried it to the living room where I was baffled to find that it would not start. I do not mean that it would not boot, but that it would not turn on the power !!!

So right now I am looking to borrow a spare power supply to see if it will then turn on the machine again. This machine with all the parts in it is less than 2 years old (about 18 months actually…) and I have started the machine maybe 4-5 times during the past winter.

Update 10.05.2007

Yesterday I connected another power supply, and successfully powered up the box. So it was indeed the power supply which for reasons unknown decided to die, between power off and power up. I have contacted the shop where I bought it to find out if I can get it fixed or replaced for free. I am not sure about that though.

16.05.2007, 22:51

Yesterday I installed KUBUNTU on my new USB pendrive, and since my other computer won’t boot from anything USB, Itriedit today on my mythtv box, and successfully booted into Kubuntu on the USB pendrive. How bout that.

Anyway, from Kubuntu I tried to connect to my wireless access point with knetwork manager, and no problem. I connected and was surfing the internet in less than a minute, including the time to select and verify a password for kwallet to hold the password to the wap.

And this was so easy after all the troubles I had with open suse 10.2, and still no wireless internet…

Well, I just tried and when I switched to knetwork managed internet, I got connected just as easy as from Kubuntu.

So my update for the mythbox, is I can now remove wireless internet from my list of things that I need to get to work. Yeah.

I am not going to do more now, since it is already late, but my mood was definitely lifted significantly just now.

And last weekend I put in some work alone, and now the boat is pretty close to being ready for the upcoming sailing season.

This could be the first year in my 4 year’s as a boat owner to actually get the boat in the water in the very beginning of april, and maybe even on april 1.st which is the day the harbour officially opens for the season.

Thanks to my crew for making this possible.

And hopefully this year we will have a long season with great sailing weather.

And not to forget, leave the other boats trailing behind in the evening matches )

Lars

An older girl (I don’t know her, and didn’t get her name but I am guessing that she was en route to a school for 8th – 10th grade) was cycling in the wrong side of the bike path and wan not looking ahead probably because it was raining and she would be getting rain in the eyes and collided frontally with my daughter. My daughter was in the right side of the path, and neither my daughter, nor the older girl reacted to my warning shout and my daughter was knocked over backwards with her bike and the older girl on top of her.

My wife whom I telephoned drove her to the dentist for a checkup because she was bleeding from lips and teeth, and she might have sustained permanent damage to one of her permanent teeth. Next week she is going for another check at the dentist.

Morale: Just abide by the traffic rules, and look where you are going. My daughter was actually doing the right thing but got hit by that girl any way because the older girl was not looking (and was driving in the wrong side to boot).