On Jun 28, 2006, at 8:25 PM, Cliff L. Biffle wrote:
> On Jun 28, 2006, at 9:49 AM, Nicholas Sinnott-Armstrong wrote:
>> I have been working on this also (Marc, if you want my new version
>> of the breakout board, LMK). I wrote a Verilog program that should
>> run the camera and attach it to an Address Bus (for external RAM
>> attachment, etc)/ Data, if you want that. This camera would be
>> amazing if we could finally get it working...
> Is the protocol documented? The cellphone cameras I've seen in the
> past were usually SPI, with an on-board compression engine...but this
> one has an awful lot of pins...
On the Spark Fun forums, there is a thread most of the way down on
the page called "Get the camera working competition!" or something
close to it. It represents the united efforts of most of us (well,
most of us are trying to win the prizes, and hence are kind of
working against each other, etc, but we all want to get it working).
We have identified the module as a HV7131GP (linked to on the first
page, but it is a dead link, try http://www.magnachip.com/ENG/
Products/CMOSImageSensor/ProductsMatrix/HV7131GP.html), which uses an
8 bit parallel interface (well, it really has a 16 bit interface, but
8 bits are brought out on to the header).
As for the compression engine, I don't think it has one, but it has a
really nice DSP processor with stuff like colorspace conversion built
in, etc, etc.
>> < /HTML>
> Is that some sort of statement? :-)
I didn't write that, it was probably caused by Apple Mail formatting
it as HTML and the mailing list daemon reading it as plaintext.
If the ARM I/O could really switch at 25Mhz or so (a little less
should work, but Absolute Minimum of 6Mhz), then it would really make
interfacing to the module a lot easier.
Hope that helps!