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

[casainho]

I have been working for some time on a portable MP3 player and came by Rockbox and your player port.  I have since switched my efforts to supporting your project instead of working on my own for now.

Good idea, I think. There are a lot of Open Hardware and firmware DAP projects on Internet, but people works alone and can't do all alone... You are welcome ;-) and we really need people that can do work, special on hardware since looks to me that Rockbox hackers are most firmware guys.
 

I am a couple of days away from sending off for a prototype based off your existing specs plus a few more for future incorporation.

I am bad on English language, can you explain better?

Looking at your daughterboard for the Olimex board, I cannot determine your switch needs and would like to discuss your design plans for user input so I can layout the board appropriately.

Well, I didn't understand all, but for user input on Rockbox Player Little, we have an expansion board designed (which I hope to get it drilled and assembled this week).

As you can see there is 8 buttons, we can buy them easily, cheap and easy to do firmware - it's what I am looking for this Rockbox Player Little "first and quick prototype".

Also, since you are using the Olimex board which is based on the Atmel evaluation board, there are a few questions that I need answered based on your current port. 

Does your bootloader use external dataflash, nandflash or just the internal dataflash?

I am thinking on putting the Atmel BootStrap + Rockbox bootloader on DataFlash - current code is working like that. But they can also be placed on the Nandflash.

Are you using or planning on using the EEPROM to persist settings?

No. Settings from Rockbox are saved on files, on FAT32 partition, where Rockbox firmware is + the audio files.

Do you see a need for a serial interface or is the USB interface sufficient?

I think serial would be good to do debug, it's cheap and simple hardware. USB is important, can be used for charge a battery, program the Atmel Bootstrap + Rockbox bootloader and transfer files to SD Card (including the Rockbox firmware). There is no working code for USB on actual port, I don't understand nothing about USB, that's why I want to use SD Card ;-) - and luckily, SD Card is very used nowadays.

So, I would just put USB + JTAG port (instead of Serial). Not everyone would use JTAG, it can be used to program the DataFlash or Nandflash, and for Debug, which is very important for hackers :-) -- Debug hardware for ARM is cheap and there is a few Open Hardware dongles (and Free Software visual Debuggers) ;-)

Do you see a need for an external RTC or use the processor to keep date/time?

Well, I don't know. I don't have any knowledge about it. I think Rockbox uses a clock and that hardware must have it...

[rasher]

It has a open source Atmel SAM7X processor, ARM7, 32-bit, 256K Flash, 64K SRAM, up to 55MHz and 48 MIPS. Ethernet, 256K EEPROM, MAKE Controller Kit.

Dear god that's slow. Why would you use such a thing?

[Llorean]

I'm sorry, could you please try to use proper English? The sentence "Your selling "open source" : a community." makes no sense (even after correcting Your to You're). We're not selling anything, and "community" has very little to do with technical requirements. That is a very slow processor. With the exception of the Archos players (which use a hardware chip to decode MP3 audio) no other Rockbox target is that slow, and a large portion of Rockbox's featureset would have to be disabled. It would more or less be pointless.

You can't make technical requirements go away by calling someone self-centered for knowing about them.

[Llorean]

We have a forum guideline asking that you make the best attempt you can to use proper English. While grammar may be difficult, spelling is something that can be checked.

Which CPU do you think is equal speed to the one you're proposing? Saying "that is considered much faster than Arm7" makes me think you're not actually looking at the rated speed, but just the architecture. The iPods are dual core Arm7 80mhz processors. That means yours is significantly slower than a single core of these dual core devices.

[casainho]

I were doing a presentation about Rockbox and Rockbox Player, at an hack lab. Everyone enjoyed it and I was invited to repeat on another place. Also there as an invitation to make a more technical presentation of internals of Rockbox Player.

I got a contact of a Portuguese company that is selling Arduino and another electronics boards like that, I hope to talk with them and who knows If I get any help......

The pictures and the presentation files are here.

You know, I got buttons working, on next days I will focus on audio - the last thing to have a full working hardware :-)

[casainho]

Hello :-)

Recently I gave a presentation of Rockbox and Rockbox Player* on
OPorto, here in Portugal. A guy there gave me the contact of a
Portuguese company (In Motion - http://inmotion.pt/store/) that sells
online Arduino boards, break boards, etc.

I contacted that company by phone and we talked a bit about the
project. I will present it next Saturday, on personal, to the project
manager of that company. From what we talked by phone, at least he
will try to help me to submit the Rockbox Player project to SparkFun -
they should have a commercial relationship.

I was told that In Motion would like to sell a project like this one,
however, they may not have money for start the project nor know how...

Wish me luck ;-)

[dkamin]

I had written a couple of weeks ago regarding an implementation of the Rockbox Player (Little) and am very close to having a final design.  I intend to offer it for sale to those looking for one to play with.  I should have the full prototype completed in the next couple of weeks and if all goes well, I will go into production shortly after.

Casainho, I would like to send you one of the prototypes when I get to that point so you can continue your development around my design.  We can talk about this in the near future.

Do you have any idea how many people are interested in getting one of the players?  They will be cheaper if I can make many rather than just a few.  I also don't want to spend money when nobody is buying.

[casainho]

I had written a couple of weeks ago regarding an implementation of the Rockbox Player (Little) and am very close to having a final design.  I intend to offer it for sale to those looking for one to play with.  I should have the full prototype completed in the next couple of weeks and if all goes well, I will go into production shortly after.

Casainho, I would like to send you one of the prototypes when I get to that point so you can continue your development around my design.  We can talk about this in the near future.

Do you have any idea how many people are interested in getting one of the players?  They will be cheaper if I can make many rather than just a few.  I also don't want to spend money when nobody is buying.

Oh, wait there!! I sent you a private message and you didn't answer or give any more explanations. Would be great if you want to work together.

First, you are not sharing any information about the hardware nor decide it with us. 2nd, if you want to work like that, you may end up alone and them you need to port the Rockbox firmware by yourself.

If you want cooperation, please start by sharing information and your ideas.

I am open to talk and work if you want to cooperate and help us and if we have the same objectives.

[dkamin]

Ouch!!!!!  I guess I deserve some of that, but not all of it.

I have not contributed before and don't know all the protocols, so be gentle.

In looking back at my original post, I see that I did not use much detail, as I assumed the basic design features were set:  MCU, memory, display, codec, etc. I am doing nothing fancy at this point.  At this time, the design is the same hardware you are working with today on the Olimex(Atmel eval kit) board with the addition of your expansion board and LED display.  I am simply repackaging it for a handheld player. 

I did not intend to deceive or withhold any information, but I can see how you came to that conclusion. 

I hope all the bad stuff is behind us now.  Let's move on to the fun stuff.


I question the need for the DATAFLASH.  It seems you can use the 4K of internal ROM on the AT91 for the bootloader rather than external ROM.  I will look further into this.  I am not designing for any nandflash unless someone sees a need.

There will be no ethernet port or serial ports.  One mini USB port and micro-SD are in in the current design.

The current design has a wheel like the sansa e200.  What are your thoughts on that?  I am not sure at this point. 

I am looking at adding either the Si4720 or Si4721 for FM support both receive and transmit.  Thoughts?
https://www.silabs.com/products/audiovideo/fmtransceivers/Pages/default.aspx

The one question remaining from earlier is related to user interface.  What are the switches purpose on your expansion board? Up/Down, Play/Stop?  Their placement depends on their purpose.

Future additions I am thinking about are GPS, Bluetooth(A2DP) and a larger display.


If there is something you would like added, let me know.


I will try to put together some docs for posting this weekend.

[kugel.]

Ouch!!!!!  I guess I deserve some of that, but not all of it.

I have not contributed before and don't know all the protocols, so be gentle.

In looking back at my original post, I see that I did not use much detail, as I assumed the basic design features were set: [...], display, [...]

I hope not! This display is crap.

The current design has a wheel like the sansa e200.  What are your thoughts on that?  I am not sure at this point. 

I am looking at adding either the Si4720 or Si4721 for FM support both receive and transmit.  Thoughts?
https://www.silabs.com/products/audiovideo/fmtransceivers/Pages/default.aspx

The one question remaining from earlier is related to user interface.  What are the switches purpose on your expansion board? Up/Down, Play/Stop?  Their placement depends on their purpose.

Future additions I am thinking about are GPS, Bluetooth(A2DP) and a larger display.

I think a wheel is controversial. Because many people dislike wheels, and it probably raises the costs too.

We, the software project Rockbox, have some drivers for Si470x, this one could be reused well. The Si4702 even features RDS afaik. But those are receive only.

And I don't think GPS or Bluetooth is needed (Rockbox doesn't have any driver for that too), but maybe a larger and better display.

[casainho]

In looking back at my original post, I see that I did not use much detail, as I assumed the basic design features were set:  MCU, memory, display, codec, etc. I am doing nothing fancy at this point.  At this time, the design is the same hardware you are working with today on the Olimex(Atmel eval kit) board with the addition of your expansion board and LED display.  I am simply repackaging it for a handheld player.

I think you should talk with us at our discussion group, all other developers are there:
http://groups.google.com/group/rockboxplayer

I question the need for the DATAFLASH.  It seems you can use the 4K of internal ROM on the AT91 for the bootloader rather than external ROM.  I will look further into this.  I am not designing for any nandflash unless someone sees a need.

On ROM there is a bootloader from Atmel, which can boots from Nand, DataFlash, USB, etc. We can NOT change that code.

For boot memories we just have 2 options, or DataFlash or Nand Flash. I prefer DataFlash, since I tested the code with it, so, there is working code for DataFlash.

There will be no ethernet port or serial ports.  One mini USB port and micro-SD are in in the current design.

Yes, it's ok. But don't forget JTAG port!!! Please, no micro-SD, just normal SD Card!! We have question to discuss here -- micro-SD don't have SPI bus, just SD Bus which mean we will need to pay royalties to use it :-( -- actual code is working for SD Card using SPI bus.

The current design has a wheel like the sansa e200.  What are your thoughts on that?  I am not sure at this point.

I don't have opinion about this. I have a Sansa E200 and I like the wheel. Let's wait for the opinions of other developers.

I am looking at adding either the Si4720 or Si4721 for FM support both receive and transmit.  Thoughts?
https://www.silabs.com/products/audiovideo/fmtransceivers/Pages/default.aspx

I will not use it, at least for this very first version, for keep it simple. I would prefer to put our energies on a better display, for example.

The one question remaining from earlier is related to user interface.  What are the switches purpose on your expansion board? Up/Down, Play/Stop?  Their placement depends on their purpose.

#define BUTTON_SELECT       0x00000001
#define BUTTON_MENU         0x00000002
#define BUTTON_PLAY         0x00000004
#define BUTTON_STOP         0x00000008

#define BUTTON_LEFT         0x00000010
#define BUTTON_RIGHT        0x00000020
#define BUTTON_UP           0x00000040
#define BUTTON_DOWN         0x00000080

That buttons, with directionals being as a cross. NOTE: that text it's a copy-paste of "button-target.h" Rockbox Player Little actual button drivers.

Future additions I am thinking about are GPS, Bluetooth(A2DP) and a larger display.

I also have some ideas, like wireless or a few inputs for recording. But first we need to get a simple working hardware and get critical mass about it, then we can try implement all our ideas.

If there is something you would like added, let me know.

Well, a better LCD ;-) Actual LCD is wired by SPI bus, I think we can change the LCD only if it's a SPI bus. This MCU have a peripheral for connect a TFT LCD, but I didn't have experience with it, so, maybe for a next version. 

Anyway, I am happy with actual LCD. And let's remind that are developers which since begin spend money and time on this project and others none, or simple don't believe on it. The ones that bought the dev. board plus all others things have merit to choose even because are the ones that will want to buy this first version.

I want to ask some questions:
- what is your motivation?
- in what can you help on the project?
- what you do professionally? do you have some projects yours that we can see?
- will you share the schematic and board layout with a Open Source license?

And more, we are talking (in discussion group) in making the expansion board, so every developer can have it - now I am the only one with it, were drilled at my work but I can't drill any more.

Others may prefer to wait for your work (I would), and in that case, how we will assembly? do you have tools? - you told about selling it, do you want to do that? I really want that someone sell to me the hardware, because I can't produce it by myself.

And what about plastic case, do you have any idea?

I received today my video digital machine - I will record the actual working prototype and post here the link to the video ;-)

[Llorean]

I'd just like to clarify that this project is really independent of Rockbox, and that any opinions in this post are those of the individuals posting. While he may think the screen is quite good, many of us developers on Rockbox disagree, and I personally am quite certain having such a small screen will limit severely who will be willing to buy such a player.

[casainho]

I'd just like to clarify that this project is really independent of Rockbox, and that any opinions in this post are those of the individuals posting. While he may think the screen is quite good, many of us developers on Rockbox disagree, and I personally am quite certain having such a small screen will limit severely who will be willing to buy such a player.

I agree with Llorean but the choice of this small screen LCD were because of this advantages on our initial situation:
    *  this LCD is used in many Open Source projects, we can get a lot of working example drivers, code and schematics;
    * this LCD connects directly to the development board being used;
    * this LCD is the cheapest colour LCD and is available in many online shops that ships to worldwide.

I must say that I did the drivers and I were alone doing it, this were the first time I worked with LCD. I took very few days to have it working, mainly thanks to quick wire to dev. board, + the tutorial from Mr. Lynch and the help of other Rockbox hackers I got on IRC ;-)

[Domonoky]

about the buttons:

It maybe good to place the select button into the middle of the directional buttons in a future hardware version. Thats the common position on all players i know.

And also more buttons cant hurt, especially if you plan to include fm and recording.  :-) 

[notlistening]

With any project ther are always those who disagree, want something different and i hear myself repeating what i have said before again and again. There are those that just complain and do nothing and those that get on with it. The most constructive comment came about the middle select button for the user interface. That has to be the best type of input as we are at the stage until we get to production where we can totally redesign this again and again so thanks for any constructive feedback. This version of the player is never going to fit in your pocket or even run off a battery. So anything you see now will not even resemble the next version etc etc.

The LCD is a sore point but the beauty of the project is that we don't like it we throw it away and bolt a new one on. But as with anything we have to learn some how and that LCD has made an almost impossible job for people who are some what novices (I speak for myself) able to get a base that they can then improve on. This is a development project and it is developing. There is a big thank you required to all those rockbox members that are helping this project along I am not going to name names you know who you are.

I want to ban any more grumbles about the LCD unless you have a suggested alternative for the next version and some feedback into the implementation of that hardware. Not just i don't like it and fold your arms.

Tom