Gabriel?: Compact Flash Problems Question

Gabriel,

I am following up on your previous thread.

http://groups.google.co.uk/groups?hl=en&lr=&ie=UTF-8&threadm=0e7c01c3af77%24c5b04b20%24a301280a%40phx.gbl&rnum=1&prev=/groups%3Fq%3Dcompact%2520flash%2520dma%2520windows%2520ce%2520problem%26hl%3Den%26lr%3D%26ie%3DUTF-8%26sa%3DN%26tab%3Dwg

I have what appears to be a very similar problem and would like to get some
feedback from you. If anyone else can provide feedback I would be grateful.

I currently have a system that is having problems with Compact Flashes.
There has been two batches of Commercial SanDisk CF that have been used: -
one works and the other doesn't. The problem is that when trying to view the
directory structure on the CF then it does not show a complete listing. Both
these CF can be fully read from a Windows 2000 system using a card reader, so
there is not a problem with the CF data itself. The problem only occurs when
used in a Windows CE system.

Do you understand the particular reason why one type of SanDisk CF worked
and the other didn't. I am trying to determine what characterises a good CF
and bad CF. I can then make sure that I am buying the correct type - not
necessarily SanDisk. I can then go to any manufacturer and get a correct
specification.

Are there any settings that I can change in the driver/registry to allow a
greater range of CF to be used?

As a general point is there anything that I can do to ensure that a Compact
Flash will work on my system? I don't want to have to go through a trial and
error phase for each type of CF. Any suggestions would be appreciated.

Thanks,
John

Dean
Thu Sep 30 07:18:13 CDT 2004

Try formatting the card on your CE device, and read it again. Maybe there's
something strange about the format on the card...

--
Dean Ramsier - eMVP


"Brian" <Brian@discussions.microsoft.com> wrote in message
news:BF67D061-2918-4301-B5D6-33B4E6A08FE5@microsoft.com...
> Gabriel,
>
> I am following up on your previous thread.
>
>
http://groups.google.co.uk/groups?hl=en&lr=&ie=UTF-8&threadm=0e7c01c3af77%24c5b04b20%24a301280a%40phx.gbl&rnum=1&prev=/groups%3Fq%3Dcompact%2520flash%2520dma%2520windows%2520ce%2520problem%26hl%3Den%26lr%3D%26ie%3DUTF-8%26sa%3DN%26tab%3Dwg
>
> I have what appears to be a very similar problem and would like to get
some
> feedback from you. If anyone else can provide feedback I would be
grateful.
>
> I currently have a system that is having problems with Compact Flashes.
> There has been two batches of Commercial SanDisk CF that have been used: -
> one works and the other doesn't. The problem is that when trying to view
the
> directory structure on the CF then it does not show a complete listing.
Both
> these CF can be fully read from a Windows 2000 system using a card reader,
so
> there is not a problem with the CF data itself. The problem only occurs
when
> used in a Windows CE system.
>
> Do you understand the particular reason why one type of SanDisk CF worked
> and the other didn't. I am trying to determine what characterises a good
CF
> and bad CF. I can then make sure that I am buying the correct type - not
> necessarily SanDisk. I can then go to any manufacturer and get a correct
> specification.
>
> Are there any settings that I can change in the driver/registry to allow a
> greater range of CF to be used?
>
> As a general point is there anything that I can do to ensure that a
Compact
> Flash will work on my system? I don't want to have to go through a trial
and
> error phase for each type of CF. Any suggestions would be appreciated.
>
> Thanks,
> John
>

Brian
Thu Sep 30 08:09:06 CDT 2004

Dean,

Thanks for the suggestion.

Just to give a little more information. This problem is not for a single
system/single CF, but a is seen on many systems with many CF. It would not be
practical to do this and the problem may occur again. So it wouldn't be
getting to the root of the problem.

Thanks,
Brian

"Dean Ramsier" wrote:

> Try formatting the card on your CE device, and read it again. Maybe there's
> something strange about the format on the card...
>
> --
> Dean Ramsier - eMVP
>
>
> "Brian" <Brian@discussions.microsoft.com> wrote in message
> news:BF67D061-2918-4301-B5D6-33B4E6A08FE5@microsoft.com...
> > Gabriel,
> >
> > I am following up on your previous thread.
> >
> >
> http://groups.google.co.uk/groups?hl=en&lr=&ie=UTF-8&threadm=0e7c01c3af77%24c5b04b20%24a301280a%40phx.gbl&rnum=1&prev=/groups%3Fq%3Dcompact%2520flash%2520dma%2520windows%2520ce%2520problem%26hl%3Den%26lr%3D%26ie%3DUTF-8%26sa%3DN%26tab%3Dwg
> >
> > I have what appears to be a very similar problem and would like to get
> some
> > feedback from you. If anyone else can provide feedback I would be
> grateful.
> >
> > I currently have a system that is having problems with Compact Flashes.
> > There has been two batches of Commercial SanDisk CF that have been used: -
> > one works and the other doesn't. The problem is that when trying to view
> the
> > directory structure on the CF then it does not show a complete listing.
> Both
> > these CF can be fully read from a Windows 2000 system using a card reader,
> so
> > there is not a problem with the CF data itself. The problem only occurs
> when
> > used in a Windows CE system.
> >
> > Do you understand the particular reason why one type of SanDisk CF worked
> > and the other didn't. I am trying to determine what characterises a good
> CF
> > and bad CF. I can then make sure that I am buying the correct type - not
> > necessarily SanDisk. I can then go to any manufacturer and get a correct
> > specification.
> >
> > Are there any settings that I can change in the driver/registry to allow a
> > greater range of CF to be used?
> >
> > As a general point is there anything that I can do to ensure that a
> Compact
> > Flash will work on my system? I don't want to have to go through a trial
> and
> > error phase for each type of CF. Any suggestions would be appreciated.
> >
> > Thanks,
> > John
> >
>
>
>

Dogbert
Thu Sep 30 08:41:40 CDT 2004


<snip>

We found we had to "tweak" some of the driver timings to get a wide a range
of CF cards to work reliably. The original driver worked OK on older cards,
but the newer faster ones required some of the time-outs increased (beyond
the official spec!).
The main problems were when the driver was trying to read
identification/status data; some of the faster cards took longer to respond.

Don't know if this is related to what you are seeing, but might be of use.

Brian
Thu Sep 30 10:35:09 CDT 2004

Dogbert,

This sounds interesting. Could you give some specifics on what would have to
change and possible value that could work.

Thanks very much.
Brian

"Dogbert" wrote:

>
> <snip>
>
> We found we had to "tweak" some of the driver timings to get a wide a range
> of CF cards to work reliably. The original driver worked OK on older cards,
> but the newer faster ones required some of the time-outs increased (beyond
> the official spec!).
> The main problems were when the driver was trying to read
> identification/status data; some of the faster cards took longer to respond.
>
> Don't know if this is related to what you are seeing, but might be of use.
>
>
>

Dogbert
Fri Oct 01 02:35:02 CDT 2004


"Brian" <Brian@discussions.microsoft.com> wrote in message
news:187A60FB-B2E4-43D5-A2DE-C1AE5AC67CE9@microsoft.com...
> Dogbert,
>
> This sounds interesting. Could you give some specifics on what would have
> to
> change and possible value that could work.
>
> Thanks very much.
> Brian
>
> "Dogbert" wrote:
>
>>
>> <snip>
>>
>> We found we had to "tweak" some of the driver timings to get a wide a
>> range
>> of CF cards to work reliably. The original driver worked OK on older
>> cards,
>> but the newer faster ones required some of the time-outs increased
>> (beyond
>> the official spec!).
>> The main problems were when the driver was trying to read
>> identification/status data; some of the faster cards took longer to
>> respond.
>>
>> Don't know if this is related to what you are seeing, but might be of
>> use.
>>
>>
>>

The main change was in the card socket driver (Socket.cin the PDD folder),
function "PDCardResetSocket" there were a number of time-outs (Sleep()) when
resetting the card and reading the card status/identity; I beleive this is
an area which is very platform specific, so line numbers (etc) will just
confuse things; the main change is immediately after a reset-off, this had
to be increased to at least 100ms, but we settled on 200 to be on the safe
side.
One or two of the others needed minor tuning to improve reliability, but
that was probably down to our platform.