CHDKPTP: S90 Primary Focal Plane Configuration - hacking out the CCD - page 66 - Creative Uses of CHDK - CHDK Forum supplierdeeply

CHDKPTP: S90 Primary Focal Plane Configuration - hacking out the CCD

  • 704 Replies
  • 114771 Views
*

Offline SticK

  • *****
  • 779
Re: CHDKPTP: S90 Primary Focal Plane Configuration - hacking out the CCD
« Reply #650 on: 14 / September / 2013, 20:48:07 »
Advertisements
PALETTE DUMP w menu up  S90_100c (good)

Full 1024 byte dumps.
« Last Edit: 14 / September / 2013, 21:22:51 by SticK »

*

Offline SticK

  • *****
  • 779
Re: CHDKPTP: S90 Primary Focal Plane Configuration - hacking out the CCD
« Reply #651 on: 14 / September / 2013, 20:49:31 »
PALETTE DUMP w menu up  S90_101a (bad)

Full 1024 byte dumps.
« Last Edit: 14 / September / 2013, 21:23:44 by SticK »

*

Offline reyalp

  • ******
  • 13387
Re: CHDKPTP: S90 Primary Focal Plane Configuration - hacking out the CCD
« Reply #652 on: 14 / September / 2013, 20:55:09 »
What does it look like when it "doesn't work"?
Figure.  On yours REC mode is fine ??
OK, this is a bug in the GUI code, doesn't actually have anything to do with rec mode or rmem. It just depends on whether the output contains % characters...

Fixed in r439 http://trac.assembla.com/chdkptp/changeset/439/trunk

As before, you don't need to update the whole build, you can just update the offending line in gui.lua. Or use CLI mode.
Don't forget what the H stands for.

*

Offline SticK

  • *****
  • 779
Re: CHDKPTP: S90 Primary Focal Plane Configuration - hacking out the CCD
« Reply #653 on: 14 / September / 2013, 21:26:45 »
Fixed in r439 ... you can just update the offending line in gui.lua.
The last post has all 1024 byte dumps.  That's the kind of fix I like ... thanks ... you're getting to +know+ me :).
« Last Edit: 14 / September / 2013, 21:38:33 by SticK »


*

Offline SticK

  • *****
  • 779
Re: CHDKPTP: S90 Primary Focal Plane Configuration - hacking out the CCD
« Reply #654 on: 14 / September / 2013, 21:37:11 »
LVDUMPS.ZIP

BIN trail ...

s90-101a-1.1.0_2592_disableisdriveerror_nodistcorr_no_startup_lens_check_isofix_trial.zip 02-23-2013 --OK
s90-101a-1.1.0-2729_disableisdriveerror_nodistcorr_no_startup_lens_check_isofix_novigncorr.zip 08-06-2013 -- OK

s90-101a-1.1.0-2729_disableisdriveerror_nodistcorr_no_startup_lens_check_isofix_novigncorr_0deg.zip 09-15-2013 - OOPS!
« Last Edit: 15 / September / 2013, 02:04:50 by SticK »

*

Offline srsa_4c

  • ******
  • 4426
Re: CHDKPTP: S90 Primary Focal Plane Configuration - hacking out the CCD
« Reply #655 on: 15 / September / 2013, 09:13:52 »
@SticK

Thanks for the dumps.
What I found so far: the palette of the "corrupted" play menu seems to be a legit palette, it's just not the right one for that menu. According to your latest update, the binary with the "0deg" hack is the problematic one. I don't see however how that could influence the palette.
It's true that the firmware addresses involved are close:
0x5d28: active palette pointer
0x5d50: orientation data
... but neither is treated as an array.

If you really want, you can make dumps on both cameras:
rmem 0x4ffc 256
1) play mode
2) play mode menu
On the 100c cam, please repeat 1) and 2) with the cam in rotated position (90 deg).

It could be, that setting that orientation-related variable introduces some confusion in the firmware...

*

Offline SticK

  • *****
  • 779
Re: CHDKPTP: S90 Primary Focal Plane Configuration - hacking out the CCD
« Reply #656 on: 15 / September / 2013, 09:54:18 »
@srsa_4c

*I* thank you for your dedication.  These are easy to do so if you need more, please ask.

edit: would you be interested in dumps of the last good version on the 101a?  Please specify what you'd like. I noticed in the previous versions before 0deg, that the 101a orientation icon did not respond in M mode, only in AUTO mode, but I haven't paid attention to it until the last couple of months since we started debugging it, so I don't know how far back it could go  ... perhaps that could be another clue, and maybe related to an earlier hack that is showing its head with 0deg ??

edit: FWIW I tested that M mode orientation icon has not responded at least all the way back to s90-101a-1.1.0_2592_disableisdriveerror_nodistcorr_no_startup_lens_check_isofix_trial.zip 02-23-2013
« Last Edit: 15 / September / 2013, 11:25:24 by SticK »

*

Offline srsa_4c

  • ******
  • 4426
Re: CHDKPTP: S90 Primary Focal Plane Configuration - hacking out the CCD
« Reply #657 on: 15 / September / 2013, 11:38:06 »
I noticed in the previous versions before 0deg, that the 101a orientation icon did not respond in M mode, only in AUTO mode, but I haven't paid attention to it until the last couple of months since we started debugging it, so I don't know how far back it could go  ... perhaps that could be another clue, and maybe related to an earlier hack that is showing its head with 0deg ??
So, it's possible that the orientation sensor is fine, and there's some memory corruption that affects various parts of RAM...? Great...

Looks like I can no longer subtract properly  :-[. The memory chunk I wanted to ask for starts at 0x5cfc (it has to cover both 0x5d28, 0x5d50). Sorry about that.
rmem 0x5cfc 256
be it
rmem 0x5c00 512
If you want, you can also try a different 101a version, perhaps the corruption is moving +/- a few words.


*

Offline SticK

  • *****
  • 779
Re: CHDKPTP: S90 Primary Focal Plane Configuration - hacking out the CCD
« Reply #658 on: 15 / September / 2013, 11:58:38 »
Oh that's interesting ... now I'm on the previous version that I showed to be OK, but (in PLAY) if I go MENU:OK, MENU: off, MENU bad! So I will get you this sequence to start with.

*

Offline SticK

  • *****
  • 779
Re: CHDKPTP: S90 Primary Focal Plane Configuration - hacking out the CCD
« Reply #659 on: 15 / September / 2013, 12:07:41 »
Hmmm ... now two previous versions that worked OK before are permanently "corrupted" .. something related to persistent storage ?
« Last Edit: 15 / September / 2013, 12:09:34 by SticK »

 

Related Topics