Rockbox Development > New Ports

Rockbox Player - Project to design and build a Free/Open hardware audio player

<< < (53/144) > >>

Davide-NYC:
Hello all,
I posted my enthusiasm for the stereo line-in *plus* toslink input jack (like the h1x0) here.

If this player does get made an optical in (for recording) would be amazing.

casainho:
I talked today with Matt about RockboxPlayerPrototypeB. He delayed it to the summer time. Also he talked something about his school project(RockboxPlayerPrototypeB) being a combination project, an oscilloscope and an mp3 player.

Since he changed the schedule of the RockboxPlayerPrototypeB, for a few months later, I will not wait for he end his project. I want to have hardware before that, to start coding - maybe in the next 4 weeks.

I would like to ask If anyone have some idea for a good route to the project. If no better idea than the last one, the RockboxPlayerPrototype, I will continue with that RockboxPlayerPrototype.

spark:
[bootloader]
I agree with Badger. We just use a resonably sized NAND chip. use a u-boot bootloader. (advangate of u boot-->we just have to do some configuration stuff and need not write one from scratch).
the u-boot bootloader will then load RB from again NAND itself.
This way things are simple.

[project route]
We should first get the stuff working on RockboxPlayerPrototype (i.e. eval board + expansion board). I can design and fabricate the expansion PCB. I'll send the PCBs across to anyone interested for assembly.

casainho, since you already have the dev board, you can start doing some coding on it. 1st step is to finalize the booting plan. i guess the above mentioned booting method is fine.

we have to start somewhere otherwise things will just keep delaying.

casainho:

--- Quote from: spark on February 24, 2008, 02:26:48 AM ---[bootloader]
I agree with Badger. We just use a resonably sized NAND chip. use a u-boot bootloader. (advangate of u boot-->we just have to do some configuration stuff and need not write one from scratch).
the u-boot bootloader will then load RB from again NAND itself.
This way things are simple.

[project route]
We should first get the stuff working on RockboxPlayerPrototype (i.e. eval board + expansion board). I can design and fabricate the expansion PCB. I'll send the PCBs across to anyone interested for assembly.

casainho, since you already have the dev board, you can start doing some coding on it. 1st step is to finalize the booting plan. i guess the above mentioned booting method is fine.

we have to start somewhere otherwise things will just keep delaying.

--- End quote ---
Okok, lets continue then with RockboxPlayerPrototype!

I didn't buy yet the dev board :-) - I just bought and have with me, the JTAG cable to program the ARM and the LCD+carrier board.

Don't worry Spark, I will make the expansion PCB, I hope to have It in the next 3 weeks. I will fabricate It and send them to developers. In the right time I will announce that and wait for request from interested developers.

I will update the wiki pages in this times and I will buy the dev board also.

I will need help to make the bootloader, I don't know nothing about u-boot :-) - Will be important to have other people doing the same work.

Project License
What should It be? - I wrote GPL2 on wiki page, however I don't know If is the better... I chose that because RB firmware uses that.

For me, a BSD would also be ok. I remember that a CC No Commercial didn't had a good reception, when Matt chose that for the RockboxPlayerPrototypeB.

I don't believe in Copyright this hardware, for example, copyright the way to wire a memory to the MCU, and since this is an hardware project, I really don't believe that If we copyright It will have any effect, other then show to others ours interests in sharing It and that anyone are free to reuse It.

spark:
[bootloader]

There will be only 1 stage of bootloading.
On power ON, the bootloader starts from NAND. The job of this bootloader is to initialize hardware and load the RB firmware into memory from somewhere (again NAND partition) and execute it.

one option is to use u-boot, the second one is to write a simple bootloader from scratch.

Getting u-boot to work will take some efforts but it does make life a lot easier in the long run. If we get u-boot to work, we can do things like updating application firmware over serial or network, creating partitions, checks, etc without using JTAGs. u-boot comes with ready NAND driver, ethernet driver, tftp boot support and filesystem support (fat,ext,jffs, etc)

u-boot is already ported for the AT91RM9200 board. we will have to use this as reference. don't worry casainho, i can give u-boot support. i have some experience with it.


[license]

I think we shouldn't worry a lot if people use it commercially or no. The more the people use it, the better it is for us.

Navigation

[0] Message Index

[#] Next page

[*] Previous page

Go to full version