In other words, how do you guys verify the settings of someone's *.inp file, only by experience (or is there another way)?

Thanks in advance! :)

-- Grand Master Peter (tehkan@powersrike.net), November 20, 2001

Answers

Yes and no. I really should wait for Ben Jos' more authoritative answer, but the basic idea is that the dips are encoded into every frame of an .inp file. Someone can look at the source for a game's driver and determine what 'proper' (default in most cases) settings would be recorded as. Then it's just a matter of verifying that the .inp in question matches up with it. The editors use a tool that assists in tournament games. For most of the games, however, stuff like that is usually spotted by experience with the game or just really obvious sort of things like having the wrong number of lives.

The MAME Olympiad movie rocks, by the way. As does Ms. Pac-Man's singing performance.

Brian McLean

-- Brian McLean (bmclean84@hotmail.com), November 20, 2001.

Thanks for the info and the comments! :P By the way, is this "assisting tool" available for everyone, or is it strictly used by the people who verify the input files?

Thanks again! (and sorry for mispelling my email address in my previous message)

-- Grand Master Peter (tehkan@powerstrike.net), November 21, 2001.

The dip checkers, slowdown analyzers and such are a jealously kept secret within the editors. :) Actually, I don't think they give it out to the general public or even us confirmers in most cases (there are two tiers of 'verifiers', the confirmers who can only say that a score plays back correctly and the editors who have the power to change other players' scores).

Brian McLean

-- Brian McLean (bmclean84@hotmail.com), November 21, 2001.

The DIP switch checker is usually updated for tournaments and distributed to all judges. It is not reserved for editors only. However, it is far from being an overall checker. Yes, Brian is correct, you can look at the source code to find out how the DIP switches are stored in the .inp. The bad parts: (1) The overall .inp format has changed at least three times, (2) there are 3000+ games in MAME, and (3) The DIP switch parts of game drivers are changed every now and then, so if you know which bit corresponds to which DIP switch for a given version of MAME, that does not mean it's the same bit in a different version of MAME, even regardless of whether the two versions use the same format as mentioned in (1) . Is anybody volunteering to MANUALLY add support for 5000+ combinations of MAME versions and games to the checker?

Also, the checker at this point is written in a way to see if the correct settings are used. Most of the time when there is a tournament, I look at what the required settings for each game in that tournament are and then manually add each game to the checker. Then it can be used to compare the settings of SOME PARTICULAR VERSION of MAME (so not just any version) of a given .inp to the required settings for that game.

So, again, it's not a one size fits all utility. Actually, it's a one size fits only a handful utility, with by far the majority being unsupported. So most of the time when I want to see which DIP switches were used in some recording, I can't even use it and have to revert to using a hex editor and reading the source code of the game driver...

Cheers, Ben Jos.

-- Ben Jos Walbeehm (walbeehm@walbeehm.com), November 21, 2001.

It is easy to show dip switch changes. for mame 0.56, go to inptport.c and add this to line 1641.

in = Machine->input_ports; in++; port = 0; while (in->type != IPT_END && port < MAX_INPUT_PORTS) { while (in->type != IPT_END && in->type != IPT_PORT) { if ((in->type & ~IPF_MASK) == IPT_DIPSWITCH_NAME) { in->default_value = (in->default_value & ~in->mask) | (input_port_value[port] & in->mask); } in++; } port++; if (in->type == IPT_PORT) in++; }

-- smame@Team2ch (smame@mameworld.net), November 22, 2001.

The idea is quite logic and finding patterns that correspond to the same or different dip switch settings of a specific game was easier than I first thought! :)

Thanks to everyone!

-- Grand Master Peter (tehkan@powerstrike.net), November 23, 2001.