well, I've gone and looked over your HEX dump(below), and although I'm not entirely sure what represents a "press" and what is a "release", I do notice the following characteristics:
1) assuming 2character hex representation:     32 ints long.   (is this 16 for "press" and 16 for "release", or just 32 on "press, or 32 on "release"?)
2) 4  clearly defined non-zero sections in each at offset 1,9,17,25 (or 0,8,16,24 + 1) -  I'll call them A,B,C,D.
3) lets focus on the int (two characters) immediately prior to each 'ff'  . eg: for the onoff, that's 7d,85,80,78. It appears that the 'D' positioned one (in this case 78) is repeated into the 'C' position on the following key press, AND the 'B' positioned one ( in this case 85) is repeated in the 'A' position on the following key press.
4) conjecture:  perhaps this "repeating D in C and B in A" is to help identify the "released" key?
5) It also tells me that since A&C are deterministic(from the previous key press), then the "relevant" part of the hex that represents the key press/release is B(85) and/or D(78), and surrounding bit/s.
good luch with fiddling with hw_hiddev.c.   
David B.
1     - 0085ff000000000000d0ff00000100000078ff0000000000002dff0000010000
2    -  00d0ff000000000000f0ff0000010000002dff0000000000000dff0000010000
3    -  00f0ff00000000000008ff0000010000000dff000000000000f5ff0000010000
4    -  0008ff00000000000050ff000001000000f5ff000000000000adff0000010000
5    -  0050ff00000000000020ff000001000000adff000000000000ddff0000010000
6    -  0020ff0000000000001dff000001000000ddff000000000000e0ff0000010000
7    -  001dff00000000000090ff000001000000e0ff0000000000006dff0000010000
8    -  0090ff0000000000006dff0000000000003dff000001000000edff0000010000
9    -  003dff00000000000010ff000001000000edff00000000000000ff0000010000
0    -  0010ff000000000000fdff00000100000000ff0000000000008dff0000010000
eax  -0070ff000000000000cdff00000100000030ff00000000000075ff0000010000
vdn -0088ff00000000000038ff000001000000c5ff00000000000045ff0000010000
vup -0038ff000000000000b8ff00000100000045ff000000000000ddff0000010000
up   -00b8ff00000000000020ff000001000000ddff000000000000e0ff0000010000
left  -0020ff0000000000001dff000001000000e0ff00000000000080ff0000010000
sel  -001dff0000000000007dff00000100000080ff000000000000adff0000010000
retn -004dff0000000000008dff00000100000070ff00000000000010ff0000010000
strt  -008dff000000000000edff00000100000010ff0000000000003dff0000010000
<missing button press here?>
cncl -00edff0000000000003dff000000000000cdff00000100000030ff0000010000
rec  -00cdff00000000000040ff00000100000030ff000000000000bdff0000010000
ff    -005dff000000000000bdff000001000000a0ff00000000000040ff0000010000

From: lirc-list-admin@lists.sourceforge.net [mailto:lirc-list-admin@lists.sourceforge.net] On Behalf Of John Porterfield
Sent: Monday, 19 December 2005 9:48 PM
To: lirc-list@lists.sourceforge.net
Subject: Re: What driver do I use for this remote

Well, irrecord doesn't quite work, well, it does and it doesn't.  I get to the part where it tells me to hold down a button and get nothing, unless I press alternating keys.  If I cat the /dev/usb/hiddev0, it doesn't show anything when I press the button more that once, it shows data for the first press and then nothing unless I press another button.  As far as editing the file for the receiver itself, I wouldn't know where to begin other than it's probably hw_hiddev.c.  I did try the latest cvs version but have the same issue.

I posted a hex dump of the output when I press the keys, but I don't know if that would be useful.
On 12/18/05, Buzz <buzz@oska.com > wrote:
Ok, well, from those pic's it;s quite obviously noting at all like the dvico model, apart from the fact that the symptoms are the same as I got when I was trying to get my DViCO remote to work initially.....
(1) it's USB
(2) it's detected by linux as a 'hiddev' device (human interface device)  - shows USB reciever hardware is OK.
(3) cat /dev/usb/hiddev0 gives binary data on remote key press  - shows remote transmitter hardware is OK
I'd try using irrecord and see if you can get it to create a config file... and if not, then get the CVS version of irrecord and try again... finially, do what I did, and hack the irrecord tool with a dogied-up version of the devinput.c  (or similar driver file)- which you'll have to mess-with yourself to get it working.   If you get it working (whick hopefully you will eventually), sent the changes and/or config file to the list and christoph, and he'll hopefully be able to add support to lircd for everyone else in the future who has one of those remotes.   :-)

From: lirc-list-admin@lists.sourceforge.net [mailto:lirc-list-admin@lists.sourceforge.net] On Behalf Of John Porterfield
Sent: Friday, 16 December 2005 11:34 PM  
To: lirc-list@lists.sourceforge.net
Subject: Re: What driver do I use for this remote

On 12/15/05, Buzz <buzz@oska.com> wrote:
message I get:
"The File You Are Looking For Is Inaccessible."

From: lirc-list-admin@lists.sourceforge.net [mailto: lirc-list-admin@lists.sourceforge.net] On Behalf Of John Porterfield
Sent: Thursday, 15 December 2005 9:29 PM
To: lirc-list@lists.sourceforge.net
Subject: [Norton AntiSpam] What driver do I use for this remote

Ack, sorry about that, you can go here http://photos.yahoo.com/ycarusbog and the pictures are on there.