Cowon D2

[andiator]

The booloader can be find on my ftp ftp://andiators.homeip.net. There is also rockbox.zip with the same revision.
The newest ("bleeding edge", build every day ) rockbox.zip can be found here: http://build.rockbox.org/dist/build-cowond2/rockbox.zip

[sikon]

Hi all! 

I just posted on FlySpray a patch that seems to solve the problem of the touchscreen driver, i've introduced an error compensation mechanism on the values read from the pcf50606 and some other stuff.
I think that the wrong values were due to some touchscreen voodoo (eg signal rebound) and does not depend on how the software is made, so the only way is to use a compensation algorithm.
It works fine for me.

Please try this fix and let me know...

I tried it and it works fine! I tested it for 5 minutes and, no errors was found! Good work.

[luck3r]

Works fine for me too.

[dhkd0407]

Hi, I'm Korean D2 rockbox user. I have a few question for d2 rockbox bootloader.

1. a DMB boot loader does not have  in ftp://andiators.homeip.net/. Where Do I get the DMB boot loader?

2. rockdoom still not working?

[andiator]

@dhkd0407
Well, I have no idea if it works on DMB D2, so there is a chance that you break your D2. Than you have to use the tcctool to recover it.

If you, despite of this warning, still want to test rockbox, than post the firware version you use and the link where I can download it, and I will make one for you.

But you are testing it AT YOUR OWN RISK.

@all
Does anybody know if it would give some problems with the DMB version?

[kfoltman]

It would - or at least it has little chance to work correctly. DMB models use 1.xx or 4.xx series firmware, while non-DMB versions use 2.xx or 3.xx.

[Yeeze]

I can install the 1.xx firmware even though I have an non-DMB D2...
So you can not use the DMB with that firmware, but the other things. So it could work

[isanggon]

I compiled bootloader for 1.58. It successfully worked with D2 dmb 8gb model.

Copy below address and paste to exploer manually. You can download compiled bootloader.

http://isanggon.mireene.com/

I will compile custom build for D2, soon.

[shotofadds]

Phew, where to start?!

@buk,

Thanks for spotting that - the fix has been committed to SVN (although I'm not entirely sure why it happened at all!).

@FranzD2,

I'll take a look at your touchscreen patch soon, although I have a few local changes in that area already that might complicate things. I also need to see if that busy-wait in the button handler can be avoided, as it could have a rather negative effect on CPU usage...

btw. You can get an indication of how much CPU the touchscreen uses by comparing results from the test_fps plugin: check the difference in results when circling the screen with your finger, against not touching the screen at all. Just add test_fps.c to apps/plugins/SOURCES if you haven't done so already.

@andiator,

Thanks for hosting these builds once again. I'll post some binaries soon (mktccboot and bootloader.bin) so that users can easily make their own patched firmware files and reduce their reliance on your server...

@everyone else,

I develop on a DAB D2, so patched 1.xx and 4.xx firmware versions do work fine. Remember that if you are not sure, you can always use tcctool to safely test if a patched firmware works before flashing it.

And Doom doesn't work yet - it's not really a priority right now...

[gevaerts]

While we're listing what now works : I got USB mostly working today.

To try it, look for the #if 0 block in config-cowond2.h and change that to #if 1

[aidy]

hi,
does having usb working have any practical features yet?

[shotofadds]

There won't be much benefit for users at this point (it's basically the same as the Cowon firmware's MSC mode, but read-only). But it will be a really helpful debugging aid to help me fix the remaining issues with the NAND driver...

[isanggon]

I found "2 lines from top on bottom" problem on lcd.
My device is dmb/dab model and I fixed this problem by editing line 80 of lcd-cowond2.c, as you said.
FREQ>>23 and 22 works fine but 24 caused bottom line problem.
I think FREQ>>23 is better choice for dmb/dab model.
 

[shotofadds]

I don't think there should be any difference between DMB/non-DMB models, since they use the same LCD. The problem is getting the correct timing when sending initialization commands to the LCD controller. If the delay is too long or too short, some of the commands do not seem to be registered, which causes this problem.

Since the CPU can either operate at 48MHz (normal) or 192MHz (boosted), the number of loops in the delay counter has to be adjusted for the CPU speed. FREQ>>23 (etc) is an attempt to scale the delay to the current CPU speed, but it's obviously still not quite correct.

Maybe something like (FREQ>>24)+2 would work? You could also try adding an extra DELAY; at the very end of the for() loop in ltv250qv_write(). Feel free to adjust any of these values and report back...

I really would much prefer to find a solution that works for everyone... 

[FranzD2]

I don't think there should be any difference between DMB/non-DMB models, since they use the same LCD. The problem is getting the correct timing when sending initialization commands to the LCD controller. If the delay is too long or too short, some of the commands do not seem to be registered, which causes this problem.

Since the CPU can either operate at 48MHz (normal) or 192MHz (boosted), the number of loops in the delay counter has to be adjusted for the CPU speed. FREQ>>23 (etc) is an attempt to scale the delay to the current CPU speed, but it's obviously still not quite correct.

Maybe something like (FREQ>>24)+2 would work? You could also try adding an extra DELAY; at the very end of the for() loop in ltv250qv_write(). Feel free to adjust any of these values and report back...

I really would much prefer to find a solution that works for everyone... 

The last value you suggested to me worked fine but since i modified the touchscreen driver this value doesn't work anymore, i now often see the two lines.
I don't know if this is due to some other mod but it seems to me that a mod in other files causes the problem in the lcd driver to reappear. How is it possible?