Since there are no other session reports for Gabriel's Chinese Chess at the time of this writing (indeed no other forum posts at all), let me say, in the grand old tradition of the Internet: First!
An overview of the rules can be found in the review of the game which I submitted to the forum, and a .pdf of the rules can be found in the Files section of this game's entry.
This game was played back in January of 1992, at a gaming weekend hosted by a member of a now-defunct postal gaming club, Knights of the Square Table. I wrote about it in my column "Anything But Chess", which was a regular feature in the club's bimonthly newsletter. Here's the annotated transcript of my game vs. Brian Garbera. Captures are indicated by an asterisk. Bonus moves are placed after a semicolon.
Tracy (Different) Brian (Same)
1. d1d3* a3c3*
2. d6d4*d2*;b5b3* f5d5*;c5e5*;c3a3*
Since I had no jumps to bonus spaces, I tried to set one up, but I overlooked the multiple jump I was setting up for Brian. 3.e3b3 is better -- it sets up two bonus jumps. He can rob me of one, but not both of them.
3. ... b6d6*d4*
4. b2b5 f2d2*b2*;a2c2*;b1d1*
Considering the strength of my reply, Brian would have been better off substituting f6d6* for his final jump.
5. a5c5*;e4e2*;e1e3*;e6e4*e2*;f3f5* d1b1
6. f6f4* d4d2
Preventing his bonus jump. The configuration in the f-column and my 11 captures prevent his other jump.
7. ... a3a2
8. f4c4 (threatens c5c3*c1*, winning) c5f5
9. a1c1 (threatens a1a3*a5*, winning) c4c3
Bonus move c5c3 forces the win 12.c2b2 e2c2 mate.
12. ... d2d1 loses to 13.a2d2.
If 11 was not the losing move, then this certainly was.
In our post-game analysis, Brian suggested 14.a2a4. Indeed, it does win one move sooner: 14. ... a5b5 15.a4a5 mate.
15.d2a2 (forced) c1a1 (forced)
A beautiful sequence of forced moves! This match is my favorite of all the matches of all the games I played that weekend.
Oh, interesting--I'll have a more detailed look next time I've got a spare day. Thanks.