On the audio issue, has this been solved? I have a 16GB 3" touchscreen 2706 based player with a speaker. One of the player's annoyances is that with sensitive earbuds, you can hear the display chirp like a ticker when it scrolls text, so I hoped that different firmware might improve the problem. (You can only hear the ticker when the display is lit.)

I installed some different firmware and the chirp went away. Also, the external speaker went away. I thought it was a good trade-off until I really listened to some songs, and realized they sounded like something in karaoke mode (which I think removes the similarities between channels, in an attempt to remove the vocals which tend to be spatially in the center of a stereo song), or like when my PC's audio drivers are misconfigured for a surround system, when I've actually only got two speakers.

Would it help to compare the two firmwares in a disassembler? If you knew where to look, it might be revealing.

I think what you're encountering is probably a feature, not a bug. ;) Is there a channel mixer somewhere in all the goodness? I think you can swap the left/right channels in "software", since there's an icon for it in my bmp0, but my player doesn't have that functionality.

I can compile the SDK to run on my player, but the buttons don't work, not to mention the touchscreen. It's not reading them at all, as it turns the backlight off and I can't wake it back up (I think this is what's happening, I suppose it could be crashing). The off-button doesn't even work, so I have to either reset it or plug in the USB to wake it back up.

I have this exact player:
Amazon.com: Pro Ebiz NEW 16gb 3" TFT Touchscreen Mp3 / Mp4 / Mp5 Player (Plays Avi/rm/rmvb/flv Without Conversion): Electronicss=electronics&qid=1255963455&sr=8-1

It's a nice player. I've had it opened and it has two visible Samsung chips but most of the electronics are covered by a large "Sony Fukushima Sew UP42367A6H" battery. The screen is an "AA003AH", which works with the FD5420 driver in the SDK. You have to change the backlight PWM frequency in hw_backlight.c, BL_Init() from 10000 to 15000 to keep the screen from giving an audible whine. I got the 15000 value from analyzing the original firmware.

Welcome, Molitor.

We have the same issue as you have with audio. Check whether this decrease of the voice happens when you plug your player to some speakers or amplifier. It works well for me if I plug it into Creative PC Speakers, it is probably the same situation with you.

I am also curious where to look at ASM to explicit the right config for our audio problem. Please send me your FW (the one with audio working) to my email: flasher86@gmail.com. It will be easier for me to compare 3 FW with 2 well working audio.

Disassembling is not a problem for me, I can identify almost every function which the SDK has, unless some new wicked structure functions has been introduced into FW by the seller.

The keyboard is also applicable for the SDK. It is easy to identify GPIO set of pins, so SDK may work for you in the future. Of course I am talking about standard buttons without some weird chip to control the whole thing.

Have you managed to modify already compiled FW by ASM? Or do you just work on the current SDK C compiler?

Ok so I see you found the few opcodes you mentioned in that Zsp_getting_started.pdf
Do you think there is really enough information out there to have the list of *all* the opcodes ? :)

When you modify the code you must calculate a new correct crc32 value or turn of crc32 checks. For crc32 code have a look at this: http://alemaxx.al.funpic.de/rk27/rkwpatch.tar.bz2

I think that using information even from Microsoft for this purpose is not a crime even for linux-fan community. The reason for this is simple, this project can OPEN the way to open source software on RK27XX players. Microsoft is a monster, but sometimes you can beat the enemy with its own... source code :D

Thanks a lot. I tried to modify the code in Hex Editor. And although I knew I haven't recalculated crc32 I thought it will work anyway (rk26XX doesn't care about crc32, there is only warning, button IGNORE, and off you go ;))
btw. Is the program you mentioned a linux console tool? I don't use linux, how to compile it into Windows console tool?

Its a simple c-program which only needs std. library functions, it should compile on nearly every system. But thats a pretty old version and I dont remember what its exactly doing. I would recommend using it as a starting point for your own special-purpose tool.

Btw, I dont know why nlite cites me and then starts talking about Microsoft???

I wonder if FSF and RedHat know that somebody is using their GPL'd products without providing source.

Compiles with Mingw but you need Admin rights on Vista and Windows 7

MEGAUPLOAD - The leading online storage and file delivery service

I am trying to create a simple base.rkw patching utility but I have had problems repairing the crc, rk26xx was easy, just a MD5 at the end of the file.

I compiled the 2 base.rkw's (with a 1 byte change)and compared them to find the crc, it seems to be the last 8 bytes of the file which changes, that would suggest crc64?

I used rkwpatch.exe in a batch file with
rkwpatch.exe BASE.RKW no.bin
pause

no.bin is just an empty 0 byte file. Managed to patch base.rkw in my original firmware to get 24 size fonts in the id3 display and it works great.

Thanks for the rkwpatch.exe tool AleMaxx, I can now mod my original firmware. I might try and create a tool now that anyone can use to patch base.rkw.