Printable Version of Topic

Click here to view this topic in its original format

Forums - Hard Drop - Tetris Community _ Strategy/Help _ Multiplayer Tetris: The Definitive Rulebook

Posted by: caffeine Dec 16 2010, 04:33 AM

DISCLAIMER: I’m not perfect! If you see any mistakes, please let me know! However, please do not claim something is incorrect unless you have tested it firsthand and are 100% sure of it (it’s very easy to mix up how many lines each attack sends in real games where not only do you have multiple attacks going on at once, but they are also subject to cancelling).

Preface

Knowing the rules can really make an impact on how we play the game. Even the small details can make a huge difference on game balance and what kind of strategy we should follow (e.g. how garbage blocking can affect combos), and boy has Tetris become ever so complicated over these past few years! So, when I tried looking up these rules online, I was surprised to find out just how difficult they are to dig up. My goal is to lay these rules out in plain sight, so multiplayer Tetris strategy is no longer so cryptic.

General rules

T-Spin Mini
In Tetris Friends, a T-Spin Mini must fulfill a set of conditions within a normal T-Spin:

In Tetris Online (Japan), a T-Spin Mini must fulfill a different set of conditions within a normal T-Spin:Back 2 Back
A B2B chain will begin with a Tetris or a T-Spin that clears a line (including T-Spin Minis). All Tetrises and T-Spins (that clear lines) excuted within the B2B chain will yield a B2B bonus. Clearing a line with something other than a T-Spin or Tetris will end the B2B chain (i.e. a Single, Double, or Triple).

Combos
Combo attacks "stack" on top of other attacks. So, in Tetris Friends, if you clear a Combo 2 as a Double, you will send a line for the Double and another line concurrently for the Combo 2 bonus.

Perfect clears
Perfect clears do not "stack" with other attacks, except for combos. In other words, a Perfect Clear executed by clearing a Tetris will not send four lines in addition to the original ten line Perfect Clear attack. (Note: there are some games where Perfect Clears stack with other attacks.)

Garbage blocking
You will not receive pending incoming garbage lines until locking a tetromino that does not clear a line (i.e. you won't receive a garbage line so long as you clear a line with your current tetromino).

Garbage cancelling
With exception to garbage blocking, you will not receive pending incoming garbage until locking your current tetromino. If, with that locked piece, you yourself send an attack, then this will reduce your incoming garbage by how many garbage lines you would’ve sent. If you would’ve sent more than your opponent, he or she will receive the difference.

Hole Change Chance
There are 10 columns, and each row of garbage will contain a hole on one of these 10 columns. In normal randomness, the hole will "change" to a new column 9/10ths of the time, which is a 90% hole change chance. Some games alter this to either make garbage easier or more difficult. For example, a 100% hole change chance means that the garbage will never line up. That means you will never be able to clear more than one garbage row at a time. Whereas, when holes line up, you can clear multiple garbage rows with a single piece (as is possible in games that have less than 100% hole change).

Change on Attack
With change on attack garbage, individual attacks are sent in groups of lined-up garbage. For example, a Tetris will send four rows of lined-up garbage. The next attack may start on a different column. In games without change on attack, garbage is distributed randomly, regardless of what type of attack you make (based on the hole change percentage).

Hurry up garbage
After enough time passes, both players will receive a solid, "unclearable" row of garbage in regular intervals. Normal, "attacking," garbage spawns on top of these solid rows. Although it will wait until your tetromino locks, hurry up garbage is not subject to garbage blocking.

All Spin
I don’t actually know the detection rules for All Spins and All Spin Minis, found in Amagami Tetris and Tetris Party Deluxe. I’ll fill this in later when someone enlightens me.

Tetris Friends; E+ mode; Last updated: 12/16/10

Garbage sent | Type of attack

0 Single
1 Double
2 Triple
4 Tetris

2 T-Spin Single
4 T-Spin Double
6 T-Spin Triple

0 T-Spin Mini Single

10 Perfect Clear

5 B2B Tetris

3 B2B T-Spin Single
5 B2B T-Spin Double
7 B2B T-Spin Triple

0 B2B T-Spin Mini Single

10? B2B Perfect Clear Tetris

0 Combo 1
1 Combo 2
1 Combo 3
1 Combo 4
2 Combo 5
2 Combo 6
3 Combo 7
3 Combo 8
4 Combo 9
4 Combo 10
4 Combo 11
5 Combo 12 or more

Note: In this game, non-"E+" modes' Combo 5 and 6 send 3 lines each and Combo 7 and 8 sends 4 lines each.

What is the chance of a garbage hole change?
100%

Is there change on attack garbage?
Yes

Is there garbage canceling?
Yes

Is there garbage blocking?
Yes

How long until hurry up garbage takes effect?
Two minutes and in five second intervals after that

Is there an All Spin Bonus?
No

Notes

The rules with question marks are my best guesses, but I am not 100% certain on them. Also, I’d like to expand this thread to other games. If you'd like to help, just copy and paste the above list and fill it out for the new game. I’ll be happy to list it here and credit you. Also, if you know of any other conditions I should list, please let me know. Although, I’d like to keep this post restricted to popular licensed online games from the “Tetris DS era” onward so there isn’t a hundred useless conditions that are the same for every game anyway.

Thank you deepcube and clincher for helping me test this stuff out! Also, thank you to Kitaru, meow, and DarthDuck!

Posted by: coolmaninsano Dec 16 2010, 04:40 AM

QUOTE(caffeine @ Dec 15 2010, 08:33 PM) *

Do(es) B2B bonus line(s) change the garbage hole column?
No?


This is a definite no.

QUOTE(caffeine @ Dec 15 2010, 08:33 PM) *

What is the interval between hurry up garbage adds?
Five seconds


5 seconds if playing at a "normal speed", but if one stalls and doesn't drop a piece, for say, 15 seconds, three lines will be added.

Posted by: Noogy Dec 16 2010, 04:46 AM

add fumens!!! biggrin.gif

Posted by: caffeine Dec 16 2010, 04:51 AM

QUOTE(coolmaninsano @ Dec 15 2010, 10:40 PM) *

5 seconds if playing at a "normal speed", but if one stalls and doesn't drop a piece, for say, 15 seconds, three lines will be added.

Thanks, but I actually already indicated this in the "General Rules" section.

Posted by: iphys Dec 16 2010, 05:17 AM

Unless they've changed things from when I tested a year ago, I don't think TF sends combo garbage on a Perfect Clear either, so I also think it would be doubtful they'd send B2B garbage if you get a B2B Tetris Perfect Clear.

Posted by: caffeine Dec 16 2010, 05:19 AM

QUOTE(iphys @ Dec 15 2010, 11:17 PM) *

Unless they've changed things from when I tested a year ago, I don't think TF sends combo garbage on a Perfect Clear either, so I also think it would be doubtful they'd send B2B garbage if you get a B2B Tetris Perfect Clear.

If Perfect Clears didn't stack with combos in the past, then they've changed it since then.

Posted by: clincher Dec 16 2010, 05:31 AM

QUOTE
What is the chance of hole change upon the new attack?
100%?

It's truly random so 90%

Posted by: caffeine Dec 16 2010, 05:42 AM

QUOTE(clincher @ Dec 15 2010, 11:31 PM) *

It's truly random so 90%

I've heard this before, but I'm not so convinced. For example, I tested many, many combos earlier today. From what I saw, every discrete combo attack resulted in a 100% hole change.

Posted by: crzy242 Dec 16 2010, 06:44 AM

QUOTE(caffeine @ Dec 16 2010, 05:42 AM) *

I've heard this before, but I'm not so convinced. For example, I tested many, many combos earlier today. From what I saw, every discrete combo attack resulted in a 100% hole change.

i have had many cases where the garbage from the next attack has appeared in the same hole. so yea, i think clinchers right.

Posted by: coolmaninsano Dec 16 2010, 07:40 AM

So basically 100% hole change but 11.1111111% for where the next whole is?

EDIT: Is it true that for the middle 8 columns the probability was 11.25% but only 5% for each of the ones on the side? (old system)

Posted by: Ravendarksky Dec 16 2010, 09:43 AM

I could be wrong but doesn't a PC send 12 lines?

Also you are stating the rules for E+ TF Multiplayer... not just TF :-)

Posted by: crzy242 Dec 16 2010, 09:52 AM

QUOTE(Ravendarksky @ Dec 16 2010, 09:43 AM) *

I could be wrong but doesn't a PC send 12 lines?

Also you are stating the rules for E+ TF Multiplayer... not just TF :-)

it sends 10 on tf. the +2 you are seeing is a combo bonus

Posted by: Rosti_LFC Dec 16 2010, 01:40 PM

QUOTE(crzy242 @ Dec 16 2010, 06:44 AM) *

i have had many cases where the garbage from the next attack has appeared in the same hole. so yea, i think clinchers right.

I'm also fairly sure that I've received garbage, cleared it, and then gotten more with the hole in the same place.

Posted by: caffeine Dec 16 2010, 01:59 PM

QUOTE(Rosti_LFC @ Dec 16 2010, 07:40 AM) *

I'm also fairly sure that I've received garbage, cleared it, and then gotten more with the hole in the same place.


What I'd like to do is have an opponent just play very slowly (no soft/hard drop, holds when the piece reaches the floor every time), and then I'll do only Doubles, and fill up his stack this way. After a few games, it should become apparent whether or not it's 100%. The only thing is that I may not have time for a couple of weeks to do this test due to the holiday.

I want to also test the B2B Perfect Clear Tetris. What I can do is start off with a regular Tetris and then build up a PC Tetris from there. It may take several trials though (getting A PC is hard, much less a non-opening PC Tetris!).

QUOTE(Ravendarksky @ Dec 16 2010, 03:43 AM) *

Also you are stating the rules for E+ TF Multiplayer... not just TF :-)

Yes, I am.

Posted by: EnFuego Dec 16 2010, 02:07 PM

QUOTE(caffeine @ Dec 16 2010, 05:42 AM) *

I've heard this before, but I'm not so convinced. For example, I tested many, many combos earlier today. From what I saw, every discrete combo attack resulted in a 100% hole change.


I am very certain that I've gotten garbage in the same hole as before with the opponent doing 2 b2b tetris. Could It be 100% different holes for combos and truly random (90%) for other types of garbage?

Posted by: caffeine Dec 16 2010, 02:15 PM

QUOTE(EnFuego @ Dec 16 2010, 08:07 AM) *

I am very certain that I've gotten garbage in the same hole as before with the opponent doing 2 b2b tetris. Could It be 100% different holes for combos and truly random (90%) for other types of garbage?

Sounds worthy of a test.

Posted by: DarkERocker Dec 16 2010, 02:38 PM

Pretty interesting topic, thanks for posting this =)
I'm gonna test some of this out and hopefully get some answers out of it ^^

Posted by: meow Dec 16 2010, 03:29 PM

Did this a while ago for TOJ, info is on the wiki.

http://harddrop.com/wiki/Tetris_Online_Japan

Posted by: caffeine Dec 16 2010, 03:39 PM

QUOTE(meow @ Dec 16 2010, 09:29 AM) *

Did this a while ago for TOJ, info is on the wiki.

http://harddrop.com/wiki/Tetris_Online_Japan


The only thing is that the combo table is different than what http://harddrop.com/forums/index.php?showtopic=2981 seems to indicate. I think new testing is in order.

Btw, Clincher tested 15 minutes worth of doubles, and it was 100% hole change every time! (We also almost got a few B2B Perfect Clear Tetrises, but no dice just yet.)

Posted by: meow Dec 16 2010, 04:25 PM

Hmmm, I think what Xeal posted is correct then because it's from the config file which Wojtek decrypted. I also tested the hole change on TOJ and it lines up some of the time.

Posted by: caffeine Dec 16 2010, 04:31 PM

QUOTE(meow @ Dec 16 2010, 10:25 AM) *

Hmmm, I think what Xeal posted is correct then because it's from the config file which Wojtek decrypted. I also tested the hole change on TOJ and it lines up some of the time.

Yeah, I have the tool, but not the decryption tool. I'd like to take values straight from that file. For the hole change testing, did you do a bunch of (non combo) doubles to an slow-playing, non-attacking opponent? How many times does it line up out of, say, 30 adds?

Posted by: Rosti_LFC Dec 16 2010, 08:34 PM

QUOTE(caffeine @ Dec 16 2010, 03:39 PM) *

Btw, Clincher tested 15 minutes worth of doubles, and it was 100% hole change every time!

Slightly more rigorous test, but have you tried it when the garbage gets cleared away too, and not just left to pile up?

Posted by: iphys Dec 17 2010, 03:03 AM

Does it count as a change on attack when you nullify incoming garbage without sending any garbage to your opponent?

Posted by: caffeine Dec 17 2010, 03:19 AM

QUOTE(Rosti_LFC @ Dec 16 2010, 02:34 PM) *

Slightly more rigorous test, but have you tried it when the garbage gets cleared away too, and not just left to pile up?


No, but good idea.

QUOTE(iphys @ Dec 16 2010, 09:03 PM) *

Does it count as a change on attack when you nullify incoming garbage without sending any garbage to your opponent?


Good question.

Posted by: chopin Dec 20 2010, 09:26 AM

Ok... so why is the first one a T-Spin Single while the second is a T-Spin Mini? I still don't understand.

Posted by: Kitaru Dec 20 2010, 10:39 AM

You forfeited your TST kick upgrade.

Mini status is retarded.

Posted by: myndzi Dec 20 2010, 12:02 PM

The second one kicks. But it has two corners open, also. I'm not sure why it counts as a T-spin at all ;p

Posted by: Kitaru Dec 20 2010, 01:11 PM

QUOTE(myndzi @ Dec 20 2010, 04:02 AM) *

The second one kicks.

So does the first. The triple kick upgrades it.

QUOTE(myndzi @ Dec 20 2010, 04:02 AM) *

But it has two corners open, also.

Look again.

Posted by: meow Dec 20 2010, 02:05 PM

just for the record, TOJ counts both of those as tspin mini. did you by any chance test this on TF, chopin? nothing is ever consistent there!

Posted by: caffeine Dec 20 2010, 02:16 PM

QUOTE(meow @ Dec 20 2010, 08:05 AM) *

just for the record, TOJ counts both of those as tspin mini. did you by any chance test this on TF, chopin? nothing is ever consistent there!

Doesn't TOJ count any Zero or Single T-Spin, whose last rotation used a kick, as a mini?

Posted by: Ravendarksky Dec 20 2010, 02:58 PM

Hey caffeine, what was the deal with the b2b PC Tetris thing you were trying out? I'm interested smile.gif

Posted by: meow Dec 20 2010, 03:00 PM

i've never seen a zero line mini on TOJ. there are some other requirements for it to be a mini.

http://tetrisconcept.net/forum/showthread.html?t=1381&page=19

Zaphod and Kitaru discussed it here. i'm not sure what they concluded.

Posted by: Kitaru Dec 20 2010, 03:25 PM

From my write-up:

QUOTE
To restate our conclusion, a t-spin is mini if there is an open corner on the point-end, it clears one line or less, or reached its final (x,y) position by way of the t-spin triple wall kick. If you count the the three conditions for a t-spin to be checked in the first place, there are six rules to evaluate a twist. It could be worse, but these rules are already severe overkill in trying to nullify EZ T-Spins. All they needed to do to detect that was add an Immobile check.

It is possible that TOJ doesn't care to distinguish between a Normal and Mini Zero. Other than that, I believe the rest of the rules should hold.

EDIT: Oh god no why oh why TOJ complicates things argh ARGH ARGH ARGH NOOOOOOOO. D:


I sincerely wish I were joking.

EDIT2: It just keeps happening...


Why did no one tell me? Why didn't you tell me about the stairs minis?

Posted by: caffeine Dec 20 2010, 03:37 PM

QUOTE(Ravendarksky @ Dec 20 2010, 08:58 AM) *

Hey caffeine, what was the deal with the b2b PC Tetris thing you were trying out? I'm interested smile.gif


I don't yet know whether or not the B2B bonus "stacks" with a perfect clear Tetris (impossible with T-Spin due to overhang).

Here's one way to test it:
1. Have opponent play super slowly: no soft/hard dropping and hold when piece reaches the floor.
2. Start off with a normal tetris to begin the B2B chain.
3. Go directly into another (B2B) Tetris, but make it a perfect clear.

Subtract opponent's garbage rows from the 4 for the initial Tetris. If there's 10 left, it doesn't "stack". I tried for about 10 minutes and got close a few times. Just needed a little luck with the order

Posted by: meow Dec 21 2010, 12:17 AM

TOJ minis conclusion as discussed on irc:

1. 1 line clears
2. T was kicked
3. cannot naturally rotate

has anyone found any minis that do not fit those rules or fulfilled those conditions without the spin being a mini?

Posted by: caffeine Dec 21 2010, 12:27 AM

QUOTE(meow @ Dec 20 2010, 06:17 PM) *

TOJ minis conclusion as discussed on irc:

1. 1 line clears
2. T was kicked
3. cannot naturally rotate

has anyone found any minis that do not fit those rules or fulfilled those conditions without the spin being a mini?

Does number three mean it's in a spot where it can't rotate without utilizing a kick?

Posted by: meow Dec 21 2010, 01:53 AM

yep so in this reposted fumen, all of the spins are minis except frame 11 where you can reach that position from a non-kick rotation


Posted by: chopin Dec 21 2010, 03:16 AM

That is EXTREMELY strange.

QUOTE(meow @ Dec 20 2010, 09:05 AM) *

just for the record, TOJ counts both of those as tspin mini. did you by any chance test this on TF, chopin? nothing is ever consistent there!



And yes, Meow. I tested the TSS, TSM on TF, I am mind blown!

Posted by: caffeine Dec 21 2010, 02:12 PM

Here's my theory behind both TOJ and TF's weird detection:

They don't want this to count as a normal T-Spin:


But at the same time they do want this to count as a normal T-Spin:


And in TF, they also want T-Spin Triple kick T-Spins to count as normal, too.

Posted by: Wojtek Dec 21 2010, 03:35 PM

oh yeah, i was about to post same thing. deal about minis is they don't want count caffeine's first fumen as t-spin, but somehow they want to keep 3-corner detection without any logical reason. most simple and logical solution would be to use immobile, but looks like they prefer to keep patching 3-corner with rules so wierd that even tetris pros here can have little problem to figure them out. And they extended 3-corner rule for all-spins (don't ask) in Amagami so you get some really wierd shit.

Posted by: caffeine Dec 21 2010, 04:24 PM

QUOTE(Wojtek @ Dec 21 2010, 09:35 AM) *

most simple and logical solution would be to use immobile

Do you mean to use immobile to detect a normal T-Spin? I might have the wrong idea about what constitutes "immobile," but wouldn't that allow these type of T-Spins:

I have no problem with it, but it seems TTC wants these to be Minis.

Does immobile mean that the piece cannot move before rotating it into a T-Spin? Because that would exclude TSTs. The wiki is unclear.

EDIT:

So 1-corner T-spins are legit under immobile? =]

Also, adding the "must clear a line" rule would seem to erase TTC's T-Spin Zero model, which they apparently want to encourage at least in single player.

Posted by: Wojtek Dec 21 2010, 05:06 PM

yeah, immobile for normal t-spins, and i see no need for minis at all. and i don't know why ttc want your first fumen to be mini, what they don't like about this spin?

QUOTE
Does immobile mean that the piece cannot move before rotating it into a T-Spin? Because that would exclude TSTs. The wiki is unclear.

no.

QUOTE
So 1-corner T-spins are legit under immobile?

yes.

what's good about immobile:
immobile is simple consisted rule, easy to understand and far more intuitive (especialy in case of all-spin).

what's good about 3-corner?

Posted by: caffeine Dec 21 2010, 05:41 PM

QUOTE(Wojtek @ Dec 21 2010, 11:06 AM) *

what's good about immobile:
immobile is simple consisted rule, easy to understand and far more intuitive (especialy in case of all-spin).

what's good about 3-corner?

Hey, I can't argue with that! The only advantage 3-corner has over immobile that I can tell is that it calls for tighter "T-Slots," since the corners in a 3-corner are nearer to the T's middle than the bounding minos in some possible immobile T-Spins. And one might argue that the whole point of rewarding a T-Spin is that it's nifty when you fit a piece into a tight spot.

Posted by: chopin Dec 21 2010, 07:14 PM

I guess it seems for TOJ that if it can be rotated as a T-Spin Single it will be counted as a Single. Otherwise it is a Mini.

Posted by: myndzi Dec 21 2010, 08:11 PM

On KoS and Nullpomino (7.5), Immobile detection, mini, and ez works like this:

If the piece rotated, is immobile, and clears a line, it's a spin.
If it kicked and doesn't clear every line it is a part of, it's a mini.
If it kicked and is not immobile, it's an EZ.

In Nullpo, EZ spins can be disabled; when enabled, they enable B2B but send no garbage on their own. EZ doesn't exist on KoS.

Mini spins discourage the kind of thing caffeine was pointing out, where it doesn't take much to make a piece immobile (though, to be honest, it's pretty impractical to try and block a piece in with the bare minimum, and there are much more ridiculous allowances with 3-point than immobile in my opinion.) This means that for maximum effect, "full" spins or natural rotations are required, and these are a rather smaller set of possible spins than minis have.

QUOTE(Kitaru @ Dec 20 2010, 01:11 PM) *

Look again.





If the kick upgraded the first, it should have upgraded the second, was my point. As well, maybe you should look again?

Posted by: Kitaru Dec 21 2010, 09:23 PM

You missed the last frame of the fumen a second time, haha.

It's the exact same end position in both twists, but one is downgraded since it wasn't the direct result of a TST kick.

Posted by: chopin Dec 21 2010, 09:40 PM

So on TF... anything that is a TST kick will be counted in full? So it's not just position and cells??? ahhh!!!

Posted by: Kitaru Dec 21 2010, 11:48 PM

Yeah, TST Promotion is kind of stupid. :/

Posted by: DarthDuck Sep 1 2011, 02:48 AM

<comparison of lines sent while in normal or E mode>

I found it helpful to memorize Caffeine's work so I know when its appropriate to cancel a combo for a tspin as well as to make opportunity cost calculations. I couldn't find anything on how many more lines combos send without the combo adjustments in E+ so here are the results from my tests with the differences highlighted in bold:

[E+]
[E or normal]

Combo one: 0
Combo one: 0

Combo two: 1 (1)
Combo two: 1 (1)

Combo three: 1 (2)
Combo three: 1 (2)

Combo four: 1 (3)
Combo four: 1 (3)

Combo five: 2 (5)
Combo five: 3 (6)

Combo six: 2 (7)
Combo six: 3 (9)

Combo seven: 3 (10)
Combo seven: 4 (13)

Combo eight: 3 (13)
Combo eight: 4 (17)

Combo nine: 4 (17)
Combo nine: 4 (21)

Combo ten: 4 (21)
Combo ten: 4 (25)

Combo eleven: 4 (25)
Combo eleven: 4 (29)

Combo twelve: 5 (30)
Combo twelve: 5 (34)

After doing these tests, it's clear that combo adjustments in E+ mode do very little to nerf combos.

Posted by: zaphod77 Oct 21 2013, 07:41 PM

QUOTE(myndzi @ Dec 21 2010, 08:11 PM) *




If the kick upgraded the first, it should have upgraded the second, was my point. As well, maybe you should look again?


Rotating or moving after performing the TST kick cancels the upgrade performed by the kick, leaving you with the mini.

Clearing two lines at once also upgrades to a full t-spin regardless.

as for all spins (in amagami) for the pieces JZSL, 4 points are chosen to correspond with the four corners used for the T piece, and the same rules are applied, except using those points instead of the Ts ones. I'm not sure what the points are.

Powered by Invision Power Board (http://www.invisionboard.com)
© Invision Power Services (http://www.invisionpower.com)