Rockbox Development > New Ports

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

<< < (49/144) > >>

mzandrew:

--- Quote from: spark on January 31, 2008, 04:18:07 AM ---Matt i appreciate your efforts on the TWiki, but i think the tone in the writing should not contain anything which indicates that one person is building it single handedly.

--- End quote ---


--- Quote from: spark on January 31, 2008, 04:30:44 AM ---Casainho, There is nothing such as "my" schematic or "your" schematic. The beauty of working together is to have "our" schematic. it doesn't matter who does the drawing. A good schematic is a result of many inputs from different people.

--- End quote ---

I've changed the instances of "I" to "we" on the wiki to be more diplomatic.  And I changed "main designer profile" to "developer profiles."

However, _I_ will be drawing the schematic and drawing the parts and producing the layout and fabricating the board out of my own pocket and assembling the board out of samples I requested and parts that I bought.  Just to be fair.

Of course, what spark means is that the basic design itself wasn't mine alone, and that's true.  I've already gotten a lot of help from others.  For starters, I read this post http://forums.rockbox.org/index.php?topic=6751.msg52891#msg52891 about 6 months ago and that's shaped what I want to do almost entirely.  Also, I checked out the datasheet for the audio codec that cashainho and others decided on for the regular RockBoxPrototype and it seems like an excellent choice, so I borrowed it.

Everyone who's contributed or plans to contribute should add their names to the "developer profiles" section at the end of the twiki page in question:
http://www.rockbox.org/twiki/bin/view/Main/RockboxPlayerPrototypeB

spark:
mzandrew, keep uploading parts of the schematic as you prepare them so that we can learn and suggest.
keep up the good work!!!

mzandrew:
I have a couple of questions for everyone that I've asked on IRC and either gotten an answer but wanted to discuss here or didn't get any answer.

First, to solve the problem of not using mobile sdram but still keeping the not-in-use power extremely low, I think that powering down the dc-dc converters that power most everything on the board (except the arm9 wakeup functionality) is a good solution.  The problem is that this is the same as turning it off.  Rockbox would have to boot up again as soon as someone presses a button.  That's fine, but if you pause your player and wait a while and the device puts itself in this low-power mode and then you hit unpause/play, you naturally expect the music to start right back up again.

From IRC, scorche implied that if you pause rockbox and then shut it down, rockbox would save where it was in the song that's playing so that when you booted back up, it would be ready to start from where it left off.  Also that Rockbox would boot up quickly enough (on a 180MHz arm9, which is approximately 2x as fast as the single used core on a sansa e200 target) so that the user wouldn't mind or even notice this happening.  Alas, my hand-me-down broken ipod first gen broke before I ever learned of Rockbox, so I've never gotten to play with it except in the ui simulator, otherwise I could probably answer some of these questions myself.

My question to this forum is first, whether I got the right message from what scorche told me, and second, when Rockbox boots up, can it be set to a mode where it won't be obvious that's it's booting back up before starting to play the music again? (I know we COULD program it to do that, but how much of a code change would this require?)  To say it another way, do any of the other targets do what I'm intending (to power off the ram so that a reboot would be required)?

spark:
I can visualize the following scenarios.

[Player is in Play mode and user presses PLAY/PAUSE button to pause]

This mode is used when user wants to pause the player for a very short time, maybe to attend a phone call or something. After a while, when he presses PLAY, he needs to have instant playback.
To implement this we should not power down the player. All we need to do is put the CPU & peripherals in low power mode and keep the SDRAM in self-refresh. (so that it's contents are preserved)
(in case the user forgets to press play for a long time, then we will have to power down the player completely as mzandrew suggested above. now the user will not mind a few seconds for startup)

[Player is in Play mode and user presses STOP/OFF button]

We need to either shut off DC/DC completely. The next boot will be a clean boot. Alternately we can save only the player state (song and position) and resume after a clean boot. (not a priority). I don't know how it is done in Rockbox?

[Player is switched OFF and USB is plugged]

Device must power up, enumerate on USB, ask for 500mA current, and start charging. It can then enter idle mode and wake up on host requests.

casainho:
About saving power, I don't know. I can't help.

I would suggest to have a message on TWiki page for people discuss here and not on TWiki page, to have it clean. Matt, can you add the link for this message and remove comments on page?

touch/click wheel
controller for proximity/touch sensing -> not good choice for a recorder/audio player. People must be able to feel the keys, to have it on pocket and put play/pause, volume, without the need to look for the display!! Also I think that it's very important for visual impaired users of RB!

Navigation

[0] Message Index

[#] Next page

[*] Previous page