I am not a very good player compared to some other players on MARP. I am very limited in the games I play, but I am dedicated. And for personal reasons, I was even a lot more dedicated during T2 than I would normally have been. You will probably see me prove that I am not a very good all-round player during the next tournament. :-)

I was very lucky with the choice of games. Even though I had never played (or even seen!) 7 out of the 10 games in this tournament until these 10 games were chosen, the other three games were the numbers 2, 3, and 4 from my top 10 votes. In addition, I had played these 3 games a lot of times right before the tournament started, so I was not rusty on any of them. The only other person who was pretty lucky with the choice of games was JSW, who got his number 1 and 2 picks in T2.

Another respect in which I was lucky with the choice of games, was that most (if not all) were pretty old. So that meant that they ran fine on my relatively slow computer: A Pentium MMX overclocked to 225 MHz (3*75, for the insiders). For all but one game (Super Sprint), I did not have to revert to playing without sound and/or use frameskip, both of which make the games at least somewhat harder.

The fact that MAME 0.34 was the version that was decided on also contributed to this, because I found that a lot of games run a lot slower on my computer using later versions of MAME compared to 0.34. I think this is very sad. Instead of MAME being optimised, it looks like the emulation speed just keeps getting worse, but that is another story altogether.

In addition, my two nearest competitors, Brian McLean and JSW both withdrew from the tournament. This was very unfortunate from a competitive point of view, and I had much rather they had stayed, but I certainly respect and understand their decision. And a little while after that, my two then-nearest competitors withdrew...

Furthermore: Time. There probably are not very many people (if any) who put in more time playing T2 games the past two months than I did. And since I, unexpectedly, was out of a job for three weeks during the tournament, I had a lot of time. For several of the games I put in the (as I call it) "DOOM Nightmare approach" (I finished most of DOOM's original 27 maps in Nightmare mode); if you try often enough, you are bound to be lucky and put up a good score a few times. And with the exception of my "chosen" 3 games (Lode Runner, Pooyan, and Scramble), I probably put in AT LEAST 50 attempts on every single game in T2. And then, I even had slightly more time than the other competitors, because after maxing out Lode Runner very early on, I had to concentrate on only 9 games instead of 10.

Finally, there are some tactics I used:

(1) I studied every game very carefully. I watched a lot of .inps. To even help and force me watch .inps, I volunteered as a judge on Kung-Fu Master, one of the games I had never played before. I also visited some web sites to read tips on games, and I talked to several people about games they were better at than I. The "forcing" also applied to the games I do not like at all; I still put in a lot of effort on those and I still elicited knowledge from others on those. For instance, I had numerous long discussions with Mark Longridge on the game I liked least in this tournament, Jr. Pacman, and despite of how much I dislike that game, it is probably the game I played most during this tournament.

(2) Position on the game. I carefully kept track of what my ranking was on each game, and I kept attacking the games for which my ranking was lowest. After all, at least in theory, more leaderboard points can be gained that way. If you are in 13th position on a game, you can, theoretically, gain 12 points by improving your score, while a 2nd place can only give you one point more at most. Towards the end of the tournament, I even wrote a little program that would take the HTML page containing all the tournament submissions, filter out all the players that did not play all 10 games, and give me the "correct" leaderboard. After all, the scores of people who did not play all 10 games do not count at all. I actually inherited the number one spot on one game this way: Pooyan. And I know I have said this before, but here it is again: Despite the fact that I knew that Brian McLean had left the tournament early on, I did not stop playing Pooyan and I still could not even come close to beating his great score. And since Brian's score is removed now, I just want to make sure that his 576,050 on Pooyan is not forgotten. My lead on Pooyan over the number 2 score may be pretty big, but Brian had an even bigger lead on it over me...

(3) "Two-point games". From a relative point of view, you can gain one additional point by overtaking (on a game) your nearest competitor on the leaderboard. If you are ahead of someone on the leaderboard, but you are second on some game while he/she is first, then taking first place on that game still only gives you one additional leaderboard point, but it increases the gap between you and him/her by two points.

(4) Maintenance. Pretty much whenever I saw that someone had overtaken me on a game, I tried to get my original position on that game back. Of course, this did not always work, but this, in combination with the previous two points, pretty much drove which game I would concentrate on for a while.

(5) Whenever I improved my score on a game, I would almost immediately upload the result. While, on the one hand, this means that everybody immediately knew which scores of mine to target, and so I did not "sandbag" any surprise attacks, this, on the other hand, also may have meant that, in some cases, I forced my competitors to stop playing the game they were concentrating on and try to do something about my most recent upload. And, after all, sandbagging can backfire as well. If you sandbag a nice score on a game, but someone else is sandbagging an even nicer score... . Another thing about immediately uploading is that if, for some reason, a submission does not get accepted, that tends to leave you enough time to try again. Some of the most recent submissions to T2 were not accepted for one reason or another, and since the tournament is over now, that means that people cannot do anything about that anymore. I submitted a total of 78 T2 recordings, which usually meant that even if a recording was not accepted, my previous recording on that game that would then be reinstated probably was good enough not to make too much of a difference.

There are several games I probably will never play again now (except for those that will reappear in future tournaments), but other games, I definitely will. And it would have been unlikely that I had found those had I not participated, so I am very glad I did. For these games, I even put in some extra effort, because if I like something a lot, I want to be good at it. Of the games I did not know before the tournament started, Rastan definitely is my favourite and that also was the game I submitted most on (15 times).

Apart from a few things that are still being worked on as we speak and it looks like they will be addressed in the next tournament, I enjoyed the tournament, so I should like to thank everyone who contributed to this. In no particular order: All the other players (after all, what good is a tournament if you are the only participant? And a victory of course means more the more players participate), the tournament judge, the regular judges, and the tournament coordinator. And last but not least, those players who very willingly shared game information with me on IRC's #marp. Thank you!

Cheers, Ben Jos.

-- Ben Jos Walbeehm (walbeehm@walbeehm.com), January 03, 2000