Strange SD-Card Behavior - page 2 - General Help and Assistance on using CHDK stable releases - CHDK Forum

Strange SD-Card Behavior

  • 17 Replies
  • 3272 Views
*

Offline srsa_4c

  • ******
  • 3919
Re: Strange SD-Card Behavior
« Reply #10 on: 20 / January / 2015, 18:14:57 »
Advertisements
Didn't work. Same error.
So much for my idea. Thanks.
Quote
The difference in the card is either the mechanical dimensions
I wouldn't think so. I bet the cards operate normally when CHDK is not loaded?
We don't currently know what's behind these "card error" issues. We get these reports occasionally, but none of us current developers have encountered it personally.
Quote
Any minor hardware/firmware error?
I'd say that a firmware issue highly unlikely. You can dump the "bad" camera's firmware for comparison if you want, but I don't think the problem is there. A small difference in hardware can likely be making a difference.

If you turn on notifications for this thread, you'll be notified when somebody has a new idea and posts it here.
« Last Edit: 20 / January / 2015, 18:27:25 by srsa_4c »

*

Offline reyalp

  • ******
  • 11920
Re: Strange SD-Card Behavior
« Reply #11 on: 20 / January / 2015, 21:03:18 »
@reyalp
How about the "idle cycle" comment in loader?
Some ports have this:
Quote
// increased from 8000 to F0000 for extra delay for sx200 maybe could be 10000 ?
The a480 has the original value (0x8000), and also pokes an MMIO before.
I don't see why it would vary between cameras. On the cameras I've tried, I've never seen any difference with this code present or not.

Since the "card locked" message blinked for the "bad" card + camera combo when the card wasn't locked, I'd lean toward a mechanical issue, but it's still pretty weird.
Don't forget what the H stands for.

Re: Strange SD-Card Behavior
« Reply #12 on: 21 / January / 2015, 03:23:08 »
On the search for the difference between both cams I tried the vers.req-methode.

Both show E18, Zoom Lens Error (at different dates; lens gear is repaired) [can I clear this, like the error codes in the car?]

Maybe one significant difference:

good cam:   Driver Info 0x70002003
bad cam:     Driver Info 0x00000000

First test to find a difference between both cards not successful, same writing speed for large files ...

Do you think reading speed of A/CHDK/MODULES/FSELECT.FLT could be the reason?
Im almost sure, that there is no mechanical problem of the switch. It works fine, only in the moment when the CHDK is loaded, there is a problem, always, even with tape on the card's switch.

Re: Strange SD-Card Behavior
« Reply #13 on: 21 / January / 2015, 10:38:04 »
Update:

When testing the cards, I formatted the "good" card. Now both cards do not work with the "bad" cam.
Maybe the card had two partitions before. Tried with different partition configurations but no success so far.


Starting CHDK with a locked two-partition card works for the "bad" cam.

« Last Edit: 21 / January / 2015, 10:42:03 by Peter Machtschuß »


*

Offline reyalp

  • ******
  • 11920
Re: Strange SD-Card Behavior
« Reply #14 on: 21 / January / 2015, 13:20:28 »
On the search for the difference between both cams I tried the vers.req-methode.

Both show E18, Zoom Lens Error (at different dates; lens gear is repaired) [can I clear this, like the error codes in the car?]
No (or at least, not that I know of). It's just a record of the last error, it doesn't indicate a problem. You can get the error even if the lens does not need repair, for example if it was blocked once when someone powered it on.

Quote
good cam:   Driver Info 0x70002003
bad cam:     Driver Info 0x00000000
I don't know what these mean, but I would guess they were some kind of health status, and 0 would be the "new" condition.
Quote
Do you think reading speed of A/CHDK/MODULES/FSELECT.FLT could be the reason?
No. CHDK just checks if the file is there or not, and the OS should report this correctly regardless of speed. Timing could be an issue if the time it takes for the OS to have the filesystem ready varies, but this check happens relatively late.
We might be able to come up with a test build that gets more detail of what is happening at that point.
Quote
When testing the cards, I formatted the "good" card. Now both cards do not work with the "bad" cam.
Maybe the card had two partitions before. Tried with different partition configurations but no success so far.
Does the "bad" cam / card work properly without CHDK? I would format the card in camera, then try shooting some pictures in continuous mode, recording a movie, looking at it in playback.
Don't forget what the H stands for.

Re: Strange SD-Card Behavior
« Reply #15 on: 21 / January / 2015, 16:57:46 »
I made some further tests using five cams and a lot of SD cards:
 
5 x A480
2 x SD Kingston 16 GB Class 4  (in the table SD1 and SD2)
2 x Micro SD Samsung / SanDisk 32 GB Class 10
3 x older SD 256 MB ... 4 GB

SD Card formatted in cam (low level), CHDK 1.3

All cams and cards are working fine without CHDK. Cam 1 & 2  and SD 1 & 2 were used for a two month timelapse with ~10k pictures each using CHDK.

Here are the results:
           
SD1
SD2
other SDDriver Info
A480 1
OK
not OK
OK
0x00000000
A480 2
OK
OK
OK
0x70002003
A480 3
OK
OK
OK
0x00000000
A480 4
OK
OK
OK
0x20002002
A480 5 not OK not OK
OK
0x00000000

My sparse conclusion:

  • only problems with one type of SD card (Kingston 16 GB, Class 4, ~5 years old)
  • no correlation to DriverInfo

*

Offline reyalp

  • ******
  • 11920
Re: Strange SD-Card Behavior
« Reply #16 on: 23 / January / 2015, 21:26:47 »
Inspired by the thread linked earlier, are you using external power supplies with these cameras?
Don't forget what the H stands for.

Re: Strange SD-Card Behavior
« Reply #17 on: 24 / January / 2015, 14:33:27 »
Inspired by the thread linked earlier, are you using external power supplies with these cameras?
Error occurs at battery operation. Even with fresh batteries, CHDK OSD display says 3.02V.
Interesting thread anyway.


 

Related Topics