[fprint] AES1660

[Vasily Khoruzhick]

On Fri, Nov 16, 2012 at 11:13 AM, Andreas <krawatten at andreas-loos.com> wrote:
> Hi Vasily,

Hi Andreas

> in "what_we_know_so_far.txt", I explicitely marked all bulks that are
> changing among different runs. As I said: The bulks you considered:

I've got this data changing, and it contains data from sensor! Just omited few
transfers from your log - and voila - it works,

looks like 0x50 cmd is "start imaging", looks like:
0x50, 0x01, 0x00, 0x01 (not sure what last 0x01 means)

0x44 is "set max transfer size" cmd,
there's "0x44, 0x02, 0x00, 0x47, 0x02" right before 583-bytes transfer
0x247 in hex is 583

0x07 is "read id" cmd, there's usb device id and bcdDevice in it, and
some extra byte
(firmware revision?)

Hopefully we don't need to crack encryption scheme :)

See attachment, it's almost yours main.cpp, just few changes... It
captures data without waiting for finger,
so place finger and then run app, it will save finger-frame.ppm as result.

This weekend I'm planning to implement AES1660 driver using this prototype...

Regards
Vasily
-------------- next part --------------
A non-text attachment was scrubbed...
Name: main.cpp
Type: text/x-c++src
Size: 46479 bytes
Desc: not available
URL: <http://lists.freedesktop.org/archives/fprint/attachments/20121116/260efbbc/attachment-0001.cpp>