Sega Tetris. Suggested special rules.

Discussion about MARP's regulation play

Moderator: BBH

Post Reply
_Zaphod_
Button Masher
Button Masher
Posts: 48
Joined: Sat Oct 21, 2006 4:57 am
Contact:

Sega Tetris. Suggested special rules.

Post by _Zaphod_ » Sun Jan 12, 2014 4:46 am

Three reasonable, and very different scoring tracks can be used.

1) Break max score ties by lines counter, higher wins. if that is tied as well, break ties by game length, shorter wins.
2) Break max score ties by lines cleared, lower number wins. this is your test of bravo ability. The player who abuses the poweron pattern best will win. Theoretical perfect play is probably not known yet. as soon as you max it out, game over yourself.
3) Same as track 2, but no bravos allowed. if you score a bravo you are disqualified. the more tetrises you score, the better your performance. as soon as you max it out, game over.

I believe track 2 is closest to how it is currently scored, but all three tracks are interesting, and i think each should be used on a version.

Also note that the B system version of sega tetris has a different poweron pattern, which has not been studied as well as the system 16 version. Because of this i recommend track 2 be used for it.

And note that only track 3 may invalidate existing scores. tracks 1 and 2 can be used with existing scores.

_Zaphod_
Button Masher
Button Masher
Posts: 48
Joined: Sat Oct 21, 2006 4:57 am
Contact:

Re: Sega Tetris. Suggested special rules.

Post by _Zaphod_ » Tue Dec 25, 2018 10:42 pm

BUMP. there are still no different tracks for this game, and too many tied 1st place scores.

Specifically, different tracks can be used to tiebreak maxouts. Which one(s) are used depends on what skill we need to measure

All system 16a/b sega tetris games are identical. EXACTLY the same. inputs from one will play back on the other with same nvram, provided they start. this is only an issue with bootleg, because it accepts coining up faster than the other ones. The bootleg has no nvram, so it always gives you the poweron pattern. This is 100% confirmed.

The two B system versions are also the same, input wise, but not the same as the system 16 versions. A replay for one can be converted to the other.

One track is to tiebreak by number of lines, higher number wins. this track tests survival. Once someone gets 999999 and 999 lines, you are at number 1. This track can be applied to existing recordings. Knowledge of the piece sequence does not matter for this track. all you have to do is survive the 2500 or so pieces required to get to 999 lines for your perfect number 1 score. Level is guaranteed to be maxed before lines are, so it does not matter.

An alternate track is lowest number of lines wins ties. This mean the run that best takes advantage of the 10x scoring bonus for bravo (or All Clear in modern tetris terms) will win. The goal is to get the maximum number of points per line. This form of play includes two forms of leeching. one form is to let the level increase by timeout, to score more points because your lines are cleared at a higher level, and the other is making bravos. all sega tetris system 16 games will give the same sequence of pieces from a clean nvram. all b system games will give the same sequence (though different from the system 16 one) from poweron. So you can abuse foreknowledge of the sequence to score well. This CANNOT be removed from the game. This track can also be applied to existing recordings.

But for those who think that track is stupid, a suggested 3rd track is this. make it so you are not allowed to score any bravos at all. Any bravo moves your submission to the other track. note that, as i said, the same input will play back on multiple versions, given an empty nvram. This still rewards you for stalling to bump up the level, and rewards line efficiency. A typical maxout not abusing bravos will hit level 99 before maxing the score. for this branch you have to watch the whole replay to validate. With this track it doesn't help so much to have knowledge of the piece sequence.

a fourth possible track is score, ties broken by framecount from start press to gameover. this would discourage stalling at low speed somewhat. But i'm not sure how to count this.

a fifth possible track is lines, with ties broken by framecount. the game that reached 999 lines fastest wins. (i'm pretty sure it's impossible to reach 999 linecount before getting 999999 points.) this would completely remove the incentive to stall, but still encourage tetrising, because a tetris takes the same time to clear as a single. Again i'm not sure how to actually do this.

Note that B system version uses a single continuous sequence that crosses games, while system16 versions generates a 1000 piece looping sequence each game. It takes 2 and a half iterations of the looping piece sequence to max out the line counter (each line uses 2 and a half pieces, 999*2.5=2497.5). For the B system version, the sequence will most likely not loop during a maxout.

I recommend the following.

One B system version on track 1.
One B system version on track 2. Most bravo lore is on the system 16 versions, so it will be neat challenge to try and abuse the B system's piece sequence.
1 system 16 version on track 1
one system 16 version on track 2 (make it the bootleg, for ease of practice).
one system 16 version on track 3.
and maybe some other versions on alternate tracks.

This should create some interesting competitions.

A further note. the game accepts input from three buttons. Cabinets did in fact have all three of them (both system 16 and b system), so inputs are allowed to have all three, and this is in fact essential for high speed play, since the game lacks a clockwise rotation.

Bloxeed can be done with either track 1 or track 2. the system 18 and c board version differ significantly. They do have poweron sequences, but the advancing garbage and the random powerups mess with this, so bravos are always okay for Bloxeed.

Flash Point also needs special scoring rules.

Score by number of levels cleared (out of 100). break ties by number of lines cleared, lower better. Score is not consistent in this game. Each level has its' best clear time so far recorded. beating this gives you a score bonus. A game that clears all 100 level is guaranteed to max out all counters.

https://www.youtube.com/watch?v=wRWrBLe0VLI

It's possible to max all counters without clearing every level. clearing all 100 levels is the true test of skill. Someone who does this is number 1.

User avatar
***PL***
Editor
Posts: 650
Joined: Mon Mar 11, 2002 9:37 pm

Re: Sega Tetris. Suggested special rules.

Post by ***PL*** » Fri Dec 28, 2018 12:08 am

It sounds like you are requesting special scoring rules as opposed to creating separate tracks. Phil would have to evaluate this request.

User avatar
Phil Lamat
Regulation Coordinator
Posts: 2014
Joined: Wed Mar 06, 2002 9:19 am

Re: Sega Tetris. Suggested special rules.

Post by Phil Lamat » Sun Dec 30, 2018 5:31 am

I will have a look ... when back from my holidays (january)

_Zaphod_
Button Masher
Button Masher
Posts: 48
Joined: Sat Oct 21, 2006 4:57 am
Contact:

Re: Sega Tetris. Suggested special rules.

Post by _Zaphod_ » Sun Dec 30, 2018 9:58 pm

yeah the games i mention need special scoring rules. the fact that there are so many identical tetris system 16 variants means that there's plenty of options for alternate tiebreaker and scoring rules. You cannot eliminate randomizer abuse from this game, because the sequence is set in stone at game start, and does not depend on anything but the seed currently stored in nvram, which only advances when a game is started (to generate the piece sequence) and at game over, to determine what animation the monkey does, if you do not skip it by starting another game fast.

The higher linecount wins system, and the no bravos system reduce the impact of memorization, while my proposed scoring system 2 (which is what players in japan actually did) means that you should abuse you knowledge of the sequence to keep clearing the field. And boy did they abuse it.

Another challenge that was done in japan was to try and build a hollow > shape on the field, with everything not of the shape filled. TGM games recognize this challenge, and will award a secret grade depending on how many lines you have filled. If we adapted this to one of the roms, then ties would be broken by lines cleared, lower is better. the full grade requires that all of row 19, except the upper left be filled, and that there be a block covering it in line 20. a z piece placed horizontally with it's upper left part in the upper left corner would qualify, once row 19 is filled.

https://harddrop.com/wiki/Secret_Grade_ ... s#Finishes examples on how a secret grade is done. I'm not entirely sure that shape is possible to make in sega tetris, and even the replay that came withe sega tetris collection showed a 9 line variant, where it stepped up every TWO columns instead of one.

Post Reply