31.03.2009 Awards & Multimedia

Heho!

The progress of the project is quite slow...

The web interface is almost done (wifi scanning,network settings, etc. already works).
WIFI is quite stable now, and working without any problems and with good signal strengh.

Ali sent the case to manufacturer and hopefully they will be able to build the wodden case.

We are also nominated for two awards:
The ITs Award of the FH Salzburg and the award at the FH Kärnten.

The initial meeting with the guys from the FH Salzburg was at the 20. of March, where we got some kick of money to support our project! Thank you a lot!!

On the 24th of April we`ll go to to Kärnten for the project präsentation for the contest =)

We updated this page a few days before and added some multimedia stuff, like a video of our presentation in the FH Salzburg (It's not that good, though i was unprepared - we were told that we wouldn't have to present anything) and some pictures from the openhouse at the HTL. I also fixed all known bugs of this page, and everything should work browser idependent and smooth.

Hopefully I'll have some time to do most of the documentation stuff, my thesis and the whole power management stuff in the easter holidays. Let's see...

Best Regards
BWA

11.03.2009 The Steps grow larger...

Hey everybody =)

WIFI WORKS!!!!!!
The Problem:
After i changed to the new 2.6.29 kernel (from 2.6.27) to get Audio to work, Wifi stopped working...
I was looking for the problem for almost two months now and thought that it must have been in the kernel!
No! I JUST USED AN OTHER SD CARD! ***** KINGSTON! *gg*
The Problem simple was a conflict (i don't exactly know what was the problem), with the SD Card on MCI Slot of the Processor and the WIFI Module on Slot two. For the first test i used a damn cheap Platinum 1GB Card and for the new kernel I used a new 2GB Kingston Card... Wifi works with the Platinum, not with the Kingston... two months of work just to come to that conclusion.

But know the xMedia Hardware is fully operable!

The Software Part is also going further some new features:
- Web Interface
- Music & Video Playback
- Webradio

But there will be some limitations due to a lack of performance of the CPU:

Video:
- MPEG1, MPEG2 (& MPEG4)
- no scaling, the video has to fit the screen
- max datarates up to 350kbit/s
- only mp2/mp3 audiotracks

Webradio / Audio:
- no ogg support (only low quality)
- no wma support

The CPU is less performant as I thought it would be, but I have to live with that...

There is also some progress with the case of the player:
Alex planed and simulated the case in the last weeks and he and Michi will travel to the company, that will hopefully manufacture the case, tomorrow to talk to guys there and to see whats possible and whats not possible.

But... It's late as ever (almost half past twelve ... and i have to get up somewhat after five o clock) ... sooo...

Good Night

15.02.2009 It's gooing further

=)

After the project started quite bad and we got over 2 months behind our timeline, we are almost back on track.

The semester holidays gave me to clean up the xRemote firmware:
- Fixed capacitive interface driver (no proximity sensing implemented yet)
- Implemented hardware accelerometer driver
- Reimplemented touch screen driver (much smoother and more accurate)
- Reworked complete user interface routines (completely event driven now)
- Added inteligent event routing
- Implemented xDimension Support (but... pssst)

I also had some time to fix some hardware issues:
- VGA - WORKING!!!!!!!! - The main problem was a change in the atmel lcdc driver firmware, atmel changed the driver, so dval wasn't enabled
- Sound - working - but Wifi not =( - Problems with the Atmel MCI driver
- Bluetooth working

And some changes in the Software:
- Full mplayer integration
- MP3 Support (only 15% CPU usage with a 192kbi/s MP3 file!!!!)
- Changes in the Menu
- Implemented xDI & Accelerometer Support

Changes in the Firmware:
- Implemented event routing between xRemote & Application Processor

I also had a few hours to create videos of the project - see "Multimedia" =)

So bye^^

28.01.2009 Yeah

Yeah!!!

It's going further, step-by-step, but it's going =)

We had a presenation on the 23. and 24. and we got the new hardware ready!

I went to multipcb near munich to pick up the new PCBs on the 22. and soldered and flashed everything till the morning of the 23rd. And i got everything working, i wanted to work =)

Hardware:

- Audio (working)

- Wireless (working!!!)

- Additional SD Card Slot (operating)

- Network (full operation)

- LCD (full operation)

- Background Light - Boost Converter (works perfect)

- Board Controller (working)

- Capacitive Wheel (fully operable)

- Bluetooth (module working properly, but still not supported by board controller firmware)

- VGA (UNTESTED!!!! and might be still buggy)

Firmware:

- U-Boot overclocked and working at 190MHz with 64Mb Ram

- Board Controller

- Serial interface Engine (working)

- Capacitive Interface (working)

- Power Management (not implemented yet)

- xRemote support (not implemented)

Driver:

- LCD Driver working properly

- Audio driver (still not running with wifi, but atmel released the new audiodriver yesterday so it should work in a few
days)

- Wireless (problems with audio)

- SPI to MMC/SD (perating but still without Card Detect)

- Serial Driver (working)

Software:

- QT Embedded (working with display flipping!!)

- Mplayer (working, without MP3 - libmad support ...)

- xMedia Software

- Main Menu (working)

- File Browser (working)

- Media Player (working ... but just really really simple, only with play and stop)

- Volume Control (working ... but simple)

- Web Browser (working ... just as demo, absolutely no user input possible)

Soo...

You might have seen the other posts below, these are copied from my own blog, where i wrote a lot about the project...

Greets

Bernhard

17.01.2009 Slowly....

Slowly...

I'm working quite hard the last days/weeks/months to put my media player together... but getting further i quite hard ... just small bugs often cost me weeks of rare time. (sorry to all my firends, that I haven't had that much time for you as I wanted to have...)

But... The second hardware revision went out to the pcb house (multipcb in germany - becauses it's much faster - but much more expensive then batchpcb...) this week and should be here soon. I got almost everything working in the prototype, only the VGA is still unusable....

I placed the VGA connector an the AD converter directly over the boost converter of the main board ... EMI problems ... it's qutie difficult to get ri of this problems, but hopefully the EMI shield placed on top of the VGA parts will help to get rid of that problems =)

Software is much more critical by now ... as the ned 2.6.27 kernel came, the atmel went unusable, the guys at atmel are working on that driver, but everything goes slow...

THE WIFI DRIVER WORKS!!!!! It cost me 5 days @ 18Hours, but finally i got 802.11g wifi on my player =)

I decided to use QT for gui programming, what worked really well so far.. the first buttons already display =)

I had one problem according to the Transformed display driver (used to flip the display, because it's mounted upside down) of QT - RGB to BGR conversion had problems...

Fixing that was really curel - every second thought of mine was "What am I doing here?" - The QT Embedded Framework is that hugh .. foremost if you have to dig that deep as modifing a lowlevel driver everything is heavily confusing.

I already surfed the internet using Webkit and played mpg/divx using mplayer =)

The second revision of the remote control (at least the hardware) is finshed, but currently untested and there is a lot of software work that needs to be done..

So...

Kernel compile finished ... hopefully CIFS will work now...

Bye

14.10.2008 School

Heho...

School started =(

The lack of time is bigger than ever before...

Everything goes quite slowly and I have to report a few drawbacks.

The boost converter of my graduation project (used for the display) smoked up several times (i had problems with the basic load). Another accident that hit me quite hard was the small LDO for the analog LCD bias that went up in smoke, because of an error in the PCB, the LDO didn't die silently it burned down everything including the whole part of the PCB.

You learn from failures! I went home turned off Skype, MSN and all that stuff and for one weekend i worked really hard and the result was worth it! I designed the best boost converter if ever seen & used =) The converter includes a current limit, a voltage limit, a limit of the duty cycle and an I2C (TWI) interface! The boost converter working at 0.1 to 1 MHz (depending on the settings) is powered by an ATTiny85 (working with 64MHz Fast Peripherial Clock) and works completely stable also if there is no system load. If you want the converter, I'll sell it to you for just a few € ^^

As I reached the final grade (the fifth), a lack of time is normal. I like to be a politician (at least in school) so I run for the job of the student representative of the school and of our department, but announced that i just want the third and the second place and I got what i want =) So I have enough influence to change the things I want, but on the other hand I don't have that much work as I had last year. Thanks to all that voted for me!

So the lessons is over in a few minutes, so let's go and get some hot chocolate =)

Greets BWA

04.09.2008 Lack of time

Hey =)

Bad news....

Holidays end in just a few days =(

9 weeks were much too short to do every i wanted to do. I got Webvirtuality stable at version 2.0, but hadn't got enough time to update ftkl.at by now.

About a week ago I finally got the PCBs from batchpcb.com and as I expected, they are perfect! Perfect copper, perfect silkscreen and perfect soldermask! Special thanks to Erik Winn from Batchpcb!

Two days after I got the boards I meet with Helmut Strasser, the best teacher at out school =) and he and Markus Gstöttner alias "The solder machine" =) helped me to solder the large chips (I don't have the equipment and the experience to solder large TQFP and QFN packages).

After I soldered a few passiv components on my xRemote PCB, I started with the firmware.

And after two days of hard work I came to the conclusion that I HATE SANDISK!!!!

One of the two µSD cards I used simpliy died after one day of unsucessfull work. The other µSD did what it wanted to, sometimes everything worked really fine, 10 minutes later - nothing, I wasn't even able to initialize the card any more - another 10 minutes later everything worked fine again. Today I went shopping and bought myself a new µSD card with 2 GB of memory for damn cheap 7€ - from Kingston. What should I say - two days of playing with a scope - for nothing, the new card works perfect! Kingston!!!!

I got one big problem with the xRemote board, the DRDY pin of the QT1106 Capacitive Sensor controller is broken (short circuit under that f******* QFN package). It cost me another day of work to get the Quantum IC working without the DRDY. Normally the DRDY reports that the chip is ready to deliver the data via SPI, after a edge of DRDY the micro has a timeframe of 540µs to start the SPI transfer. What do you do without that? Answer: High speed polling - just poll fast enough that you get into that timeframe - in my case I use a 5kHz polling rate and everything works fine. But the main problem with this method is the CPU usage - about 40% of a totally overclock ATmega2561 just to read the capacitive sensors, but It work =)

Much work waiting for me in the next weeks, months, .......

Bye and good night!

04.09.2008 Holidays

Hey^^

The school year ended and the summer holidays are here.

In the first 2 1/2 weeks of the holidays I did (do) a lot of hardware stuff. I developed and layouted my next years graduation project (the first prototype), the xRemote (a small Remote Control with a touchscreen, capacitive sensors, bluetooth and ir), the xSwitch (a quite powerfull - if it works =) - audio crosspoint switch with a 4 x 4 matrix and an IR interface).

I'm going to submit the designs for manufacturing at www.batchpcb.com, the PCB service of sparkfun.com (the greatest electronic shop i know =) ). Batchpcb is damn cheap, only 2.5$ per inch² for a doublesided PCB with two soldermasks and two silkscreens.

Two days ago I found my old quad stereo audio amp on my shelf and played around a bit.

The sound quality was quite poor so i tried to improve it.

I used a wrong footprint for the audio jacks(i mixed up the left channel and gnd)!!!

After I corrected that, ohh the sound got better =)

But the sound was not that clean as i wanted it to be. I used a 20V / 4A switching supply, that produced lots of noise on gnd and the voltage output. I looked into by electronic box and saw them, two 680µH Inductors!

Yeah! Placed in line on the power supply they reduced the noise almost completely and the sound quality of the amplifier now increased drastically.

Operation done!

So I've quite a lot to do in the next days^^

Good night!

01.05.2008 The first ideas

For the past weeks I found out some very interesting things about capacitive sensors, to create wheels like the ones used in the Apple iPod, i ordered a few QT1106 from Quantum (Atmel) and some breakoutboards to test these things. If everything works fine I'll be able to create buttons that work not only when pressed directly, but also if a finger comes near them. The really great advantage is, that you don't have care about the casing that much, you simply place sensors behind the casing and you don't need to incorporate buttons into the casing.

I continued working on my remote and managed increase the performance dramatically , so that the remote is able to read small graphics fast enough to be for example be able to scroll trough them (iPod style =) ).

I'm looking forward to next years project work, I'll force myself to work with a Atmel's monster, the AP7000 CPU at a clock of 200MHz, the display of the sony psp, a wireless lan module & capacitive sensors. But more later =)

Good night