SanDisk Sansa c200v2, m200v4, clipv1, clipv2, clip+, and fuzev2

[sko]

My player played 25 songs (mp3) until now without any freezes or other problems with mmu + caches... playback seems stable, but I'll let it play more songs to be sure. 

[jago]

I found that with forcing asynchronous bus-mode my radio is working again (Actual SVN).

Code: [Select]

#if 1 // (AS3525_FCLK_FREQ % AS3525_PCLK_FREQ)
#define ASYNCHRONOUS_BUS                          /* Boosted mode asynchronous */
#endif

EDIT: radio is working again (was not working with svn)

Yes, lowering the DRAM_FREQ to 31 MHz is working too.

[stray]

FS#10048 (newest patch) does not work on my Fuze.
Test_disk/test_fps runs fine. No white screens,screen corruption etc.
But sometimes my scrollwheel stops working, and instead it controls the backlight. Oo

/edit: Building the database also didn't work.

[matsch]

By the way some GPIO pins settings were removed since they had no effect (at least on the Clips tested so far). You could try to add them back (one by one and then the 2 of them)

Just add these lines at the top of lcd_init_device() after ams3525_dbop_init()

Tried to add the GPIO pins setting back. Without success. Working Screen with very weak contrast for my Clip.

[Hillshum]

Tried to add the GPIO pins setting back. Without success. Working Screen with very weak contrast for my Clip.

As in you weren't able to get the settings back in or that worked but didn't help the problem?

[matsch]

Tried to add the GPIO pins setting back. Without success. Working Screen with very weak contrast for my Clip.

As in you weren't able to get the settings back in or that worked but didn't help the problem?

I compiled with GPIO pins settings back, but this did not help.

The DC-DC converter is disabled in lcd-ssd1303.c.  Does OF also disable the DC-DC converter?
Perhaps my clip OLED-Driver has to run with converter.

[funman]

The DC-DC converter is disabled in lcd-ssd1303.c.
Does OF also disable the DC-DC converter?

Yes the init routine is the same than the OF.

You can try enabling the as3514 dcdc like Fuze & e200:

Code: [Select]

ascodec_write(AS3514_DCDC15, X); /* current = X * 1.25mA : ranges from 0 to 0x1f = 38.75mA */

[matsch]

you can try enabling the as3514 dcdc like Fuze & e200:

Code: [Select]

ascodec_write(AS3514_DCDC15, X); /* current = X * 1.25mA : ranges from 0 to 0x1f = 38.75mA */

OLED Driver of my Clip is supplied by as3514 DCDC. Screen is now bright with code above and X = 0x10. Will test other values for X. Thanks for the tip.

[funman]

Cool! Please post a patch to FlySpray when you have something clean.

Also note if it makes any lightning difference with different values.

[fuflo]

OLED Driver of my Clip is supplied by as3514 DCDC. Screen is now bright with code above and X = 0x10. Will test other values for X. Thanks for the tip.

i can confirm this as well. i also have european clip v1 without FM. i'm not much of a developer, especially with this low level coding, but i can help test patches

[FlynDice]

I posted FS#10270  dbop_button_lcd_delays.patch which adds some dbop related delays into the e200v2 Button & LCD drivers.  I believe the lcd delay fixes the "white screen" failures we've been seeing and the button delay will hopefully fix the spacerocks bug at low pclk settings.  I will accept a big "I told you so" from kugel on the button delay even though he was nice enough not to say so...    I believe the lcd delay is not present in the e200v2 OF but is in the fuze OF lcd driver.

[funman]

OLED Driver of my Clip is supplied by as3514 DCDC. Screen is now bright with code above and X = 0x10. Will test other values for X. Thanks for the tip.

i can confirm this as well. i also have european clip v1 without FM. i'm not much of a developer, especially with this low level coding, but i can help test patches

Can you confirm that you have no radio chip (radio not detected by rockbox) ?
You can check in debug menu -> FM radio.

Perhaps this is the way to detect the Clips which needs this special screen powering.

[fuflo]

Can you confirm that you have no radio chip (radio not detected by rockbox) ?
You can check in debug menu -> FM radio.
Perhaps this is the way to detect the Clips which needs this special screen powering.

yeah, debug menu says "hw detected: no"

[funman]

I thought that all Sansa AMS had a FM radio chip built in, and that it was disabled by software.
Probably because it was simpler to have only one board design, and so they avoided to pay taxes for devices using FM reception.

Sandisk released "different" firmware upgrades by providing identical files with a one letter extension indicating region and FM chip presence.

Recently they released firmwares for "all regions" (at least 0x.01.32 , last Clip firmware available today).

So I suppose (without proof) that they built newer boards without a FM chip, and that on these new boards they changed the way the LCD screen is powered. My 2 disassembled clips are "ver 1.5", and the Clip disassembled by anythingbutipod is "ver 1.3".

Note : the OF (verified v29, 30, 32) sets the DCDC15 register to 1, in its lcd_enable() implementation, so if we can't identify models which need this powering we would only "waste" 1.25mA on models which don't need it.

@fuflo & matsch : is your LCD functional when setting the register to 1 ? Is there any visible difference with higher values (up to 31 == 0x1f) ?

If you have disassembled your Clip (if you didn't, no need to open it), what's the hardware revision printed on the board? (between the home button and the directional keypad)

[fuflo]

well i tried 0x1, 0x11 and 0x1f and didn't see any visual difference, works with all of them.

my clip was manufactured in 2007 but i don't know what's the revision of it.

the serial/model no. on the box is: SDMX11N-2048K-E70, maybe that helps