Menu

Show posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

Show posts Menu

Messages - mcht

#1
Tetris / Toward E-sport
June 21, 2019, 07:56:52 AM
[ What makes a game interesting ]
1. it's fun and exciting during gameplay
2. some players are better than others, for a reason

[ Modern Tetris ]
Modern tetris is all for speed, like almost all other games else. But, there are two kinds of speed.

The first kind of speed is operatioin speed. There are many different causes that can slow down your operation speed, for instance, your finger may not lift up from a key fast enough for another finger to start. If you have really swift fingers, they also help you play piano and guitar beautifully.

The second kind of speed is thinking speed. Thinking speed is not such a straidforward thing as operation speed. Modern tetris is all about pattern recognition. Let's say there are two players A and B. As for a simple pattern (e.g. place I block into a ditch), A and B can place the pieces with the same speed. But for a complicataed pattern (e.g. a surface potentially for a TST setup), A could come up with the correct placement much faster than B, because A has established a circuit in the brain for that pattern. Without that circuit, B simply can't recognize the pattern, or has to step-by-step calculate out the following placements.

The really interesting thing is how operation speed and thinking speed interact. Suppose a player can complete an operation within 500ms. If it takes him more than 500ms, his talent for operation speed is just wasted. It's like typing, when you type a key, the next key to type is in your mind already. The operation and thinking run parallel to speed up. As a result, a tetris player will choose a strategy that suits best with his talent. If he is very good at operation speed that he can place 3 pieces in a second, he should play really fast to overpace the opponent. If he is good at pattern recognition, he should slow down to maybe 2 pieces per second and play with really good decisions.


[ puyo puyo tetris ]

puyo puyo tetris might be the best modern tetris game so far. The game design emphasize on "efficiency". But efficiency is actually the synonym for "brain speed". Below are listed how ppt tries make players play brain speed, not operation speed (refer to http://harddrop.com/wiki/Puyo_Puyo_Tetris) :

. Long line clear latency.
. No garbage blocking. This basically restrict combo.
. Not always clean gargage for a tetris or t-spin. This is very important that it in a way impacts opponent's efficiency.

Moreover, modern tetris actually is designed for efficient placement. 2 lines clera for only 1 garbage line, the efficiency rate is 1/2. tsd has efficiency rate 2. You do something difficult (you think more with a little more milliseconds), you get your reward.

[ Problem of ppt ]

The problem of ppt is obvious: the strongest player is not the most efficient one. This may be a little controversial, but I think Japanese players kazu (カズ) is the most efficient player in the world. Maybe even the game designer of ppt is surprised by kazu's play:

. kazu plays incredible back-to-back games, he can come up with very complicated t-spin setups within a really short time. He seldoms stop to think. For optimal efficiency, you play perfect clear (if you are a robot) or back-to-back.  
. kazu always try to make the surface very flat. When he stack up for a tetris, the result is flat surface. And playing t-spin is intrinsically aiming for a flat surface, because T-block fills an irregular hole.
. kazu doesn't just attack, when the surface is not flat or bad for continuity (e.g. can't place s/z block), he give up t-spin and try to stack up for tetris, or just do some line clears.
. The worst nightmare of kazu's opponent is kazu attack and defend at the same time. He always tries to reveal the hole of the first underlying garbage line. So even though the attacking is fierce, it's still possible for him to handle the counter-attacking. Using garbage lines to do back-to-back is the ultimate technique of pursueing efficiency.

But the sum of above is still not enough to defeat amemiya (I said it will be controversial. He did it once, this february, but with no luck in other matches).

Amemeya has been the ppt world champion for many years. About 4 years ago he won over hebomai. If you watched the match, the reason he won must be quite obvious to you. hebomai was playing ppt the way he played TF and TOP, but ppt awards attacking style. As far as I know ppt is the only tetris game amemiya plays, and he plays it really a lot. Amemiya is pragmatic, that is, he only chooses the techniques that guarantee his wins. Fast, bright, precise, playing whatever works, positional and defensive, like Capablanca.

Note that I am not going to debate what's the best style or who will be the best player in the world here. What I am interested is how to improve the game design. What I care about is efficiency.

When a player is in danger (almost being topped out), of course he should try anything to defend. As for ppt, he should use b2b to defend because that's the most efficient way. But there are several reasons that it's a little bit difficult to do:

. The situation is quite dangerous, so it's risky to do a tetris, though if the tetris is made, it's really helpful for defense.
. Sometimes there is not much space up there, not even possible to do a tetris or there t-spin setups.
. t-spin chances are not always availble
. garbage lines have already come...
. potential timing attack

So plain line clear is the best choice sometimes. Line clear also can help improve the surface to make following t-spin setup or tetris stack-up easier. What I want to say is use plain line clear to defend is not a problem at all. It has to be played after all.

The problem is use garbage lines to attack. I believe this is something ppt game design team didn't pay a lot attention to. amemiya actaully plays a lot of this during his games:

1. Opening:

I wrote a simple article about how to defend combo opening:
http://harddrop.com/forums/index.php?showtopic=8166

If you follow the opening systems I mention in the article, I believe it's totally okay for you to play a combo player who is as fast as you in the opening. But there is an exception. Sometimes the garbage lines your t-spins send will be used by the opponent for his last few combos. Let's say the height of the opening stack-up is 10 (60 squares, 15 pieces, ~2 bags), thus 9 combos, the situation is pretty even, because you can handle the 20 garbages lines he starts to send after your makes two t-spins. But if the holes of garbage lines you send out are very handy to use for combo (of course you can't control that), the garbage lines back to you will be 24 / 29 / 34 / ... no one can survive that, even a t-spin god like kazu.

The opening amemiya loves very much is a variation of DT cannon that turns into 4w whenever playable. The two t-spins it sends don't kill the opponent, the following 4w does.

2. Mid-game after opening:

amemiya plays plain line clears that sometimes reverse the game. Like said in the beginning, a game needs to be fun / exciting. If a player can get out of bad situation, that's exciting. But if he suddenly reverse the game, that might look a little bit too much to me. A player who is in bad situation does need some materials to get out, thus the hole of the garbage lines. But it should not be a hollywood-type plot twist. If the player stack up on his own and start to clear the lines all the way and send a lot of garbage lines to defend or even counter-attack, totally okay with that, no problem. But using the holes of the underlying garbage lines to make combos, is , out of control. It becomes a random disaster. After watching amemiya and kazu's midgames, it's not difficult to see the different strategies they take:

. kazu tends to adjust the position with t-spin and tetris, that is a very efficient way to play (this is no secret among b2b players but not easy to do though), while amemiya may directly choose plain line clear for the same purpose.

. kazu loves spike / TST / DT cannon / double tsd / STSD / ... , in order to create large amount of garbage lines that is impossible to counteract. amemiya can do those setups very well too, but he also loves to stack up a bit and see what happen then. There is a chance the holes of garbage line are perfectly positioned for combo. Not only it's quite safe to play like that, but also can be deadly. Of course the b2b setups are very good for efficiency.

My personal opinions for the game design is: if a player take low efficiency strategy, that's a problem of game design.


[ Improvements over ppt ]

Solution 1: garbage line combos produce no garbage lines.

The garbage line combos can still be used for defense, they just don't attack. This eliminates the possiblity of killing your opponent with combos generated by garbage lines.

Solution 2: b2b with garbage line produce extra garbage lines

It's more difficult to play b2b with the hole of garbage line than creating a b2b all by ourselves. The hole is so dynamic that players choose to combo with it for a reason. If garbage lines prduced by b2b change from 1 to 2, player will love to play b2b with the hole of garbage lines.

These 2 solutions mean different garbage line systems for regular lines and garbage lines, that may conflict with guideline.  
#2
Strategy/Help / Albatross vs Hachispin?
January 22, 2019, 04:23:32 AM
Quote from: Skyfire
Hi, I'm a somewhat high-ranking player (approximately 2700th) in Puyo Puyo Tetris, and with early O T-spin openings, I've been either using Albatross or Hachispin, which both seem like effective openers. However, I'm not sure when either of these setups is better than the other. Other than the need for an early O piece, what are the pros and cons of either setup, and in what situations are they best used in?

My puzzle league ranking is among top 200 (PPT PC). Basically I do both, the reason is for speed:
1. If O-block is the first piece, it can be just placed at the same location. No need to decide which orientation of opening to play at the first piece. It's straightforward then.
2. Actually I know only a few players playing "mirrored" openings. If mirrorred opening is played, every placement of it has to be learnt and played for two orientations, that would take a relatively long time. At least for the first and second bags it won't take long to master both.

Albatross might be the most efficient opening, producing timely tsd at the first bag, and sub-timely tst at the second bag (if early I-block is handled perfectly. As for speed, in practice it's a little faster than hachipin. Though hachipin has a good-to-continue surface after second bag or third bag. And hachipin also has stable output (not much to think about stack-up). I can't really say which one is better.
#3
Quote from: Okey_Dokey
In the first post, we have basically all the standard openers. Problem is 4-wide performs a little better at the top level (in clients like Puyo Puyo Tetris, Tetris Friends or Tetris Online Poland). It's a matter of luck: The non-combo player has to send garbage quickly which results in clean garbage that the combo player often can use to enhance his/her combo. For the same reason, I like to start with a T-Spin Single (Mr. T-Spin's STD; Hachispin works, too) - to not give the combo player a free Tetris.

Let me explain further here, the "system"/"solution" I mentioned in the article only includes :
  • 1 effective openings that works well under combo opener's attack. Maybe old school, but still good choices.
  • 2 100% complementary openings, that is, considering all combinations, there shouldn't be a chance you can't do nothing but just wait for overwhelming garbage lines. Even the best PC opening player encounter this kind of situation when there is just no PC for some combinations. (though I still put PC in solution 1). When you just play fst from the beginning, sometimes there are simply no more than 1 b2b within the first 3 bags.
  • 3 least openings which are necessary. 4 for solution 2 and 3, 6 for solution 4 (actually it can be 5).
Regarding to the garbage lines used for combo, actually I am planning to write an article about it. It's a little subtle: I intentionally avoid talking about it in this article, but the article is a preparation for it.

#4
Thanks for the great input. I agree that if there is a relatively complicated opening like BT cannon that can send tsts and also certain chance for PC, it would be a great weapon against c4w.
#5
Hello everyone, today I am going to share with you the tetris openings I use to play against combo openings. This may sound a little bit unfriendly to combo players, right? But as a back-to-back player there is a reason why I have to think more about openings, that is, I lose really a lot of games to combo players, including 4w players, c4w players, and puyo puyo players. I have been playing puyo puyo tetris PC version for about half a year and got a lot of trouble playing with combo openings. The common result of combo opening, played by either side, is a short game, very very short game. I may win unexpectedly, or lose miserably. The videos I make are basically an attempt for survival or even some lucky quick wins.

By the way, as you can read, I am not a native English speaker, but since I am going to work with foreigners in the near future, some English language practice might be good to me. So...  please bear with me.

Now let me start with my personal rule of thumb playing against combo openings: I have to be fast and precise. That sounds useless huh? Yeah totally useless. Okay let me put it another way: 2 tsd or tst in my opening. I have to do two tsd or tst, fast and precisely, in my every opening. Why t-spins? Because tetris is slow, taking roughly two times longer than t-spin. And Why better with tsd or tst? Because that's is the best I can do so I simply have to try to avoid tspin-single. Let's review how frighteningly strong combo openings can be. As for 4w and c4w, the first combo, or say the second line clear, produces zero garbage line. Then, the second combo can produce one garbage line. Not very impressive huh. But if the opponent manage to make 9 combos, what I got now are 20 garbage lines, occupying the entire 20x10 region. But I may still survive, because I could defend by t-spins or other line clear techniques. Basically the game is even when the 4w player makes 9 combos, suppose our operation speed is really close. But if he or she makes 10 combos or 11 combos, there would be 24 and 29 garbage lines sent to me respectively, then the situation is really bad to me. Therefore I need to play fast and precisely with 2 tsd or tst from the beginning, and better with a flat surface for easier defense later.

And what about puyo puyo? Almost all puyo players can play very very strong opening. Basically puyo players play combo through the game, but their opening combo is obviously the strongest. For extreme cases, some puyo players are so strong in the opening that you will find the playing very much relaxing, after you are out of the opening. Games with them are like opening, relaxing, opening, relaxing, ... . The calculation depth of puyo players is really amazing that I have a hard time coming out with 4 or 5 puyo clears in a row. The best strategy for tetris player may be continual harrassment, that is simply to break the perfect chain of puyo. So it's critical to send them many cute little stones from the beginning, otherwise just overwhelmed by massive garbage lines, that's all.

It's important to know that this video is definitely not a tutorial or introduction, but just an experience sharing or sorting out, from a tetris lover. The openings mentioned will be relatively easy and doable, therefore I am still playing them.

Tetris openings for playing against combo (2)

<<< TKI >>>

TKI is by far the most common b2b opening. It's:
. Straightforward, so you can do it fast
. The conditions are easy to meet. TKI opening succeeds if (not only-if) I-block comes as the first or second block.
. Lots of variations
. A flat surface after 2 t-spins.

Below I list the variations I often use. The classification is based on the placement of L-block.

[ L-block on top ]

[fumen]http://harddrop.com/fumen/?v115@vhDBQJKpBEmBGhQSAMoo2APG98AQ+brDFbEcEP5BAA?[/fumen]

It looks natural and it might be the fastest opening as for the first tspin. There are at least 3 sub-variations for the second one.

[fumen]http://harddrop.com/fumen/?m115@vhJBQJKpBEmBGhQSAMoo2APG98AQ+brDFbEcEP5BAA?vsBDtBFqBvpBEhBAAPAA[/fumen]

If you are good enough with tspin, this variation basically takes you into the mid-game immediately. You can happily try to tspin all the way at the left side until you are out of ideas.

[fumen]http://harddrop.com/fumen/?m115@vhOBQJKpBEmBGhQSAMoo2APG98AQ+brDFbEcEP5BAA?vsBDtBFqBvpBEhBpoBjsBinBFlBBsBmpB[/fumen]
I strongly suggest TKI perfect clears, because there is hardly any penalty with this attempt, you have tspins plus perfect clear, or you have tspins alone. When I say "no penalty" I mean you can play it without slowing down at all. Perfect clear might be the only way you could anti-attack against combo openings, normally all you do is just defend. When you play b2b opening with another b2b player, you simply play for some humble advantage over each other. However, combo openings are aimed for finishing the game before entering mid-game, not just for advantage. As for 4w and c4w, one of the weakness they share is they have to downstack line by line, that takes time before combos start to take effects. To an experienced combo opening player, perfect clear that acompanies with tspins is the only real threat to that weakness.

[fumen]http://harddrop.com/fumen/?m115@vhNBQJKpBEmBGhQSAMoo2APG98AQ+brDFbEcEP5BAA?vsBDtBFqB+tBvpBaqBMjBRmBHZBtsBRhE8YeAAA[/fumen]

This one is strong, but it needs the second L-block to come early, and the hole at the left side to be filled.

[ L-block at the right side -1 ]

[fumen]http://harddrop.com/fumen/?m115@vhDBQJKpBEmBWyB[/fumen]

If Z-block is not there, L-block can be placed there. A no-brainer.

[fumen]http://harddrop.com/fumen/?m115@vhHBQJKpBEmBWyBPnBDoBFqBAAA[/fumen]

A very simple opening, providing 2 solid tsds.

[ L-block at the ride side - 2 ]

[fumen]http://harddrop.com/fumen/?m115@vhOBQJKpBEmBvsBDtBFqBWyB0pB3lBDoBCjB2kBBgB?lrBzrB[/fumen]
[fumen]http://harddrop.com/fumen/?m115@vhOBQJKpBEmBvsBDtBFqBWyB+oQMAzYANEO388AQz7?xDslQAAvpBaqBjiBxfBlrB7rB[/fumen]

This might be one of the best opening. After you place the L-block at the right side, you may quickly go for tspin setup because it's apparently just the kind of surface for it. Or, you may go for perfect clear and tspin at the same time as shown in above two fumens. Other placements towards perfect clears are also possible.

[ L-block against the wall ]

[fumen]http://harddrop.com/fumen/?m115@vhFBQJKpBEmBvsBTtB+jB[/fumen]

This can be used when O-block comes earlier than L-block.

[fumen]http://harddrop.com/fumen/?m115@vhIBQJKpBEmBvsBTtB+jBFqBMjBvrBHhD8FeD8GeB8?QeAAAvhBlsBAAA[/fumen]

This one is also quite strong, the surface is not very flat though. It might not seem so but the small pool at the right side is really good material for tspin, you place an L-block for a tsd, or place L-block and J-block for 2 tsd, or Z-block for a tsd and a tst afterward.

I am not very into long variations, two reasons:
1. They don't work always. Of course when it works the number of garbage lines would be the optimal, but I can't count on that. Fun stuff though.
2. There are too many possible combinations, sometimes the long variation actually works, but it takes time to find the ordering of placements. Need to pursuing the ultimate speed in the opening.

<<< The Effectiveness of TKI Against Combo > >>

4w: Effectively stop the stack up. Because TKI normally sends 2 tsds, that's 9 lines, 4w opening stops at height 11, that's 10 combos, 24 lines.

c4w: same as 4w, effectively stop the stack up.

puyo puyo: it's just like playing mid-game against puyo puyo when TKI is chosen. Just try my best to send them garbage lines constantly. The side who is faster at setups will win.

<<< DT Cannon >>>

To me DT cannon is all about LZ pair and JS pair.

[fumen]http://harddrop.com/fumen/?m115@vhI2QJyuB0mB3kBplBTtB1hB6jBOYB[/fumen]
[fumen]http://harddrop.com/fumen/?m115@vhIPMJKrBeqBMpBplBTtB1hB6jBOYB[/fumen]

When LZ pair is placed or JS pair is placed before other blocks, DT cannon works.  Actually, solely L or J block as the first or second block could guarantee that DT cannon is doable (as shown in second fumen). Personally I don't like to "mirror" the opening, that is, I always make tspins at one specific side. So I don't have to get used to do it at the other side.

The first T-block is not used for tspin so it could be a little bit slow if the second or third T-block come late. DT cannon is amazing:
1. One tsd plus one tst, 11 garbage lines! When it's done very fast, basically combo opening doesn't succeed, that is, the problem the title claims is fixed.
2. Though it looks a little messed up at the left side, it's always possible to stack up in a short time, if you are skillful.
3. It is not so difficult to meet the LZ/JS condition.
4. Strong following-up variations, which are all based the the 3 squares of the 4 columns at the right side. Trying to analyze a bit, what could we do now:
a) 4w. But what are facing is ongoing combo attack, so this may be not a good idea.
b) Make a tetris. Slow and weak but this is easiest.
c) Sometimes I do another one set of a tst and a tsd (or two tsts), the key is S/Z placement:

[fumen]http://harddrop.com/fumen/?m115@vhX2QJkmBplByuBDtBnkB1hBOYB6jBJaBXXBlnBjkB?kaBvbBtsBEmB+tB6eBDfBxgBHiBdoBAAA[/fumen]

d) tspins plus tetris or tspins plus perfect clear
[fumen]http://harddrop.com/fumen/?m115@vhc2QJkmBplByuBDtBnkB1hBOYB6jBJaBraBHXBlnB?sbBDTBkaB6XBvRBmVBtsBRcBetBCWB3SBMeBJVBlYBCbBpo?B[/fumen]
[fumen]http://harddrop.com/fumen/?m115@vhY2QJkmBplByuBDtBnkB1hBOYB6jBJaBraBHXBlnB?sbBDTBkaB6XBvRBmVBtsBRcBLaBSyBMtBFtB[/fumen]

c) and d) need a lot of practice. The problem with DT cannon and all openings that use T-block for stacking up is the complexity of the accumulated pieces. TKI generally has the first tspin upon the first bag, and the second tspin upon the second bag, quite timely. DT cannon can have at most 13 pieces not cleared that when you have such amount of pieces to be dealt with you think there got to be something big and strong ahead to compensate the serious result the tspin delay causes, but after the first 2 bags actually it's kinda out of control, the ideally prepared setup may not come true that there is a possibility you always have 13 pieces stacked up.

Normally variation d) is not so bad, because:
1. There have been 2 tspins anyway.
2. Perfect clear is surely big and strong.
3. Its hit possibility is high enough to take a shot: 1. SZ placements of the second bag 2. then one of I-block and L-block doesn't come early (they depends on J-block and S-block respectively).
4. tspin is another choice if L-block of the third bag comes earlier than S-block.
5. Even the setup fails at last, you can still go for a tetris, just back to what can be done in (b).
6. The surface is quite flat at left side. Prepare the tspin setup before I-block comes.

But if the opponent plays combo, at the third bag of tetriminos, garbage lines may start coming, So perfect clear is out of question at late stage. Better do it at the second bag of tetriminoes. There are also second DT-cannon variations and other perfect clear variations.

<<< The Effectiveness of DT Cannon Against Combo > >>

4w: DT cannon is the main reason 4w players stop stacking up at height 9~10.

c4w: Effective sometimes. c4w players actually have a good time with DT cannon, when it's not done fast enough. The worst case is the end of 3 bags of tetrominos. Of course sometimes it can be done at the beginning of the third bag, of that 4w can't take the risk, but c4w can.

puyo puyo: Effective, but I doubt it can do better than TKI. Intuitively tst might do the harassment job better than tsd, due to the weakness of puyo puyo, the unexpected massive blocking of the chain. When the blocking is being cleared on the way, the positional irregularness also accumulates. But actually it's not exactly like that. puyo puyo players are good at holding a chain and release with perfect timing. When the chain is potentially being blocked, there are techniques to avoid or clear the blocking. The chain is still there even being blocked. The best strategy playing against puyo puyo is constant harrassment, becuase the length of chain is the resource of the power. The long chain is forcibly given up while the speed of tetris kicks in. No accumulation, no power. The keyword here is "constantly", so it's quite important to make the first tspin and second tspin timely. First tspin of DT cannon is available when L-block and J-block of second bag have come. And once there is a T-block in hand it's better to use it to clear lines immediately (always try to decide variation or other stuff during line clear). No bad feelings, puyo players, this is just an effort for better games, and I could be all wrong.


<<< TD Cannon >>>

It's safe to say TKI and DT cannon together have solved 70-80 percent of problems. Here is yet another opening that may increase the percentage to 90%. You will find it quite useful for some impossible tetrimino combinations. TD cannon is complementary to DT cannon, in the sense that it goes with LS/JZ pair instead of LZ/JS pair, as follows:

[fumen]http://harddrop.com/fumen/?m115@vhD+NJXsBFiB6eB9gE8EeF8DeF8DeE8OeAAAvhBdoB?AAA[/fumen]

It starts with a tst, that is rare among various openings. And as you can see the second tspin can be a tsd or tst. If it's a tst, there could be 13 garbage lines sent. But its drawbacks are also obvious. There is a lot of empty space at the left side that takes time and technique to fill. I don't know any regular placement for the second bag. The only rule of thumb I follow is to place the I-block at the bottom and it will be fine. As a result, the first tspin of TD cannon is quite slow. As far as the second tspin, I only do tst when it's going to be a quick one, that is, pieces are handy to build tst setup (e.g. O-block plus S-block or L-block plus J-block or many other combinations). It's even better when there are two S-blocks coming so I can have 2 tsds that send garbage lines fast and steadily.

TD cannon can be played at only one side you like, but I find it's very easy and useful if played mirrorred. Maybe this is because I have never learnt any variations of TD cannon.

What so good about TD cannon is its ability handling early Z-block and S-block, though this may be not good enough. When an early Z/S-block appears, it's natural to search for the corresponding L/J block ahead. If L/J is then placed before the hold space is used up, TD cannon setup works. TD cannon fits the mindset for Z/S block that TKI and DT cannon can't provide, because in general Z/S-block are not placed at the bottom while TKI or DT cannon are the chosen openings. Let's take a look at how powerful TD cannon can be, if the first blocks are Z->S->T, the direction of TD cannon will be decided by the ordering of L-block and J-block. Just place ZST stack with the right direction then wait for other pieces in the first bag. Another example is T->S->I->L, though you have "good" pieces I and L in the sight, you simply can't use TKI and DT cannon for this permutation. Even you can use TKI and DT cannon for some special permutations, that may not be executed intuitively. The good thing about opening is you don't need to calculate too much.

Finally, is there any combination we can't handle so far with TKI/DT/TD? Yes, it's the T-block and O-block as the first two blocks. T-block depends on S/Z block and O-block depends on I-block (actually you can directly place the O-block at the bottom in some cases if you are capable of deeper calculation). There are 3 openings I find useful for T-O combination.

<<< The Effectiveness of TD Cannon Against Combo > >>

4w: TD cannon is a nightmare to 4w players.

c4w: Due to its essential slowness I suggest tsd as the second tspin but not tst. Easy math: you send extra 2 lines, but each of extra c4w stackup lines send 4 or 5 lines while it's more than 7 combos.

puyo: Puyo could have accumulated a long chain before the first tspin setup of TD cannon. So maybe it's also good to have another tst instead of simply tsd, the idea is to aggresively block the chain and let the chain and the garbages occupy the space and limit the development. Kinda risky play. Need a flat surface to continue the attack otherwise it's meaningless.


<<< Perfect Clear >>>

Perfect clear is very much fun to play. There are two kinds of playing in modern tetris, one is spatial pattern recognization, the other is direct calculation. Perfect clear belongs to the latter (just take it as I don't know what I am talking about). I think practicing perfect clear helps develop tetris sense. Generally perfect clear opening is like this:

[fumen]http://harddrop.com/fumen/?m115@vhH+NJjnBKsBhiBNpBkfBHqBAAA[/fumen]

If T-block comes before Z or S, the first bag of perfect clear can be well placed (just place the I/L/J blocks at the corner in a low-energy state). Then the first 4 pieces of second bag will decide if the 10 lines of perfect clear can be sent out or not. Perfect clear opening might be the most efficient opening that it sends 10-12 lines with only 10 pieces. Moreover, this amount of garbage stops all kinds of combo setup most of the time. And combo opening from the other side won't disturb you because it's done within 2 bags of tetriminoes.  And I bet you the remaining surface is by far the flattest one in the world. The only problem with perfect clear opening is: it doesn't always work. For very strong perfect clear players, the success rate can be 70-80%. But for other players it can be like 50-60% or even lower. Perfect clear has its own system, that can be played independently of all other openings. BT cannon is another great system there to be explored.

When perfect clear opening doesn't work, what is the player supposed to do? The best strategy is to build up a spike. Soft-drop a L block or J block to fill the underlying hole then try a tspin setup at the central column. But it will take time because the surfaces of the left side and right side of the spike are not at the same row. Now here is the situation: the first back-to-back is delayed (you may have softdrop at least twice and I-block is late), and the second one is delayed too. There are merely tetris and tsd, thus 9 garbage lines at length. If you are playing with a b2b player, that's not so bad, just try to come back in the mid-game; but as for combo, that's a dead or alive thing, your comeback chance is nowhere to find. I mean, this is unacceptable because the player can only rely on possibilities but can't do a thing about it. Please note that to perfect clear experts the benefits may outweigh the risks. And, it's fun and exciting. So, I am just saying perfect clear opening might not be for everyone. Above simply explains why I prefer other openings than perfect clear opening (over the T-O combination): because I am a lousy perfect clear opening player. Thus I need to find something to replace it with.


<<< The Effectiveness of TD Cannon Against Combo > >>

When perfect clear opening succeeds, of course it's very effective. The second perfect clear that clears two lines only is deadly.


Below listed is our solution 1 to play against combo:

TKI opening +
DT cannon opening +
TD cannon opening +
perfect clear opening



<<< Hanging Openings(?) >>>

The two openings that needs only O-block as the first block and send 9 and 11 garbage lines, respectively.

<A>
[fumen]http://harddrop.com/fumen/?m115@vhIrLJMtBnhBmqBClBJkBlnBOiB6tBHhF8EeD8FeC8?PeAAA[/fumen]

<B>
[fumen]http://harddrop.com/fumen/?m115@vhOrLJHtBshBCqBmlBJkBlnB3cB6tBOaB5dBkgBDeB?tkBAAA[/fumen]

The name of A is "albatross". Don't know the name of B. There are 3 ways to play them:
a) use A only, mirror it if necessary
b) use B only, mirror it if necessary
c) use A and B, always place O-block at the same position

I use c). Firstly let'scheck A, it might be the most efficient b2b opening. It's not hard to fill the flat left side. If I-block of the second bag comes early, the tst won't be completed til the third bag. I think there are some variatons that follows but I never learn them. When I-block and T-block of second bag both comes earlier I just try to make a flat surface.

And opening B looks weird but actually it's good. After the tst, you can have another tst or just a tsd, very easy to setup. The tss is a little bit weak but the later tst and tsd(tst) will compensate.

No matter which one of a) & b) & c) is chosen, you need to know that if L-block comes earlier or J-block comes early. If you can't foresee either of them, then just place S-block and Z-block as you like, it will work.

Like said both openings need only early O-block for its setup thus meets the requirement of our T-O combination problem. Now here comes another question: is there an opening that needs only T-block? Yes, there is.

<<< The Effectiveness of Hanging Opening Against Combo > >>

Maybe not the best openings used to play against combo, comared with TKI and DT cannon, but, they still work. I still play them. A can be really fast and strong when I-block comes late, and B is also great when there is no soft-drop at all. For instance, when it's L-O or J-O combination, I play A and B immediately.


Below listed is our solution 2 to play against combo:

TKI opening +
DT cannon opening +
TD cannon opening +
hanging opening(s)



<<< Pyramid Opening(?) >>>

I admit it's a stupid name, but I don't know if there is a name for it:

[fumen]http://harddrop.com/fumen/?m115@vhGVQJMqBPsBihBupBbtBJkB[/fumen]

This T-block supports other three blocks like a boss. Simply hard-drop the T-block and the opening can be done. I can't find a regular placement of the second bag, but in practice it's quite easy.

[fumen]http://harddrop.com/fumen/?m115@vhMVQJMqBPsBihBupBbtBJkB6jBOYBvbBkaBLXBpEB?[/fumen]

If I-block comes as the first block of second bag, it needs to be held but T-block may come early too, thus regularness of placement is gone. Anyway, it's not difficult at all to place the pieces of second bag.

This opening can fix the drawback of hanging opening: you don't have to look far away for the L-block or J-block.

<<< The Effectiveness of Pyramid Opening Against Combo > >>

Similiar to DT cannon. Note that it can have an early tsd too, as long as L-block and J-block of the second bag are placed.  


Below listed is our solution 3 to play against combo:

TKI opening +
DT cannon opening +
TD cannon opening +
Pyramid opening


In next section, another approach is taken to find an even simpler/stupier opening solution.

---------------------------------------------------------------------------------------------------------------

What if we could instantly come up with which opening to play at the first sight of foreseen pieces in the beginning of a game? If it's possible, how many pieces do we need to foresee?

Let's try one piece at the first place:

[ 1-piece ]

I-block: TKI always works, hard-drop it
L/J-block: DT cannon always works, hard-drop it
O-block: hanging openings, hard-drop it
T-block: pyramid opening, hard-drop it
S/Z-block: unknown to me, maybe such kind of opening does exist. In some cases DT cannon just doesn't work, in some other cases pyramid just doesn't work.

It's a open question to me if one piece is possible. But two pieces and three pieces is possible, as shown below. We will just list the combinations and doable opening to each of them.

[ 2-piece ]

Most of the time it might work, because we can just hold S/Z block and hard-drop the next block for its corresponding doable opening. But we can choose openings that are more intuitive for the combinations since there are 2 pieces foreseen now.

IL: TKI
IJ: TKI
IZ: TKI
IS: TKI
IO: TKI
IT: TKI

LJ: DT cannon
LZ: DT cannon
LS: TD cannon
LO: hanging openings
LT: pyramid

JZ: TD cannon
JS: DT cannon
JO: hanging openings
JT: pyramid

ZS: unknown.
ZO: hanging openings -> ?
ZT: pyramid

SO: hanging openings -> ?
ST: pyramid

OT: pyramid

Problems:
1. SZ is pretty tough. Still have no idea what openings to use without the knowledge of later pieces.
2. The solution for ZO and SO is not good enough, though I know hanging openings are feasible, I still have to know which of L-block and J-block comes first otherwise I don't know how to place Z/S. The so-called legendary "far far away LJ" problem. Here is a solution for it:

[fumen]http://harddrop.com/fumen/?m115@vhODNJXrBMsBehBKpBpeBFqBOnBkkBpmBLgBKUBnaB?doBAAA[/fumen]

This is called STD (tss+tst+tsd) as I know. I really love this opening, it's ingenious. It always works while OS or OZ are the first two pieces (I and L/J block depend on them). Moreover, when you play this opening with a b2b player, sometimes the hole of the received garbage line may ruin the opening by appearing below the I-block of second bag. Combo player can't ruin it.

Now come to the stage of 3-piece... maybe all openings have their good and bad, what is more important is if it's possible to fit in a system. If it can find a complementary opening there, if it can fix a specific combination all others can't, then we may say it's good, for the system.


[ 3-piece ]

There are 35 combinations, each of that has 6 permutations. I just put the most intuitive opening for each. I have to say it's pretty up to choice of player.

< I >

ILJ: DT cannon
ILZ: TKI
ILS: TKI, but for the permutation S->L->I if you play only one-sided TKI and this is not your preferred side, TD cannon is better.
ILO: TKI
ILT: TKI
IJZ: TKI, but for the permutation Z->J->I if you play only one-sided TKI and this is not your preferred side, TD cannon is better.
IJS: TKI
IJO: TKI
IJT: TKI
IZS: TKI
IZO: TKI
IZT: TKI, but for the permutation T->Z->I if you play only one-sided TKI and this is not your preferred side, TD cannon is better.
ISO: TKI
IST: TKI, but for the permutation T->S->I if you play only one-sided TKI and this is not your preferred side, TD cannon is better.
IOT: TKI

< L >

LJZ: DT cannon
LJS: DT cannon
LJO: DT cannon
LJT: DT cannon
LZS: DT cannon, but for the permutation S->Z->L if you play only one-sided DT cannon and this is not your preferred side, TD cannon is better.
LZO: DT cannon
LZT: DT cannon, but for the permutation T->Z->L, both pyramid and reversed DT cannon has to be played. I prefer pyramid.
LSO: TD cannon
LST: TD cannon
LOT: hanging opening

< J >

JZS: DT cannon
JZO: TD cannon
JZT: TD cannon
JSO: DT cannon
JST: DT cannon
JOT: hanging opening

< ZSOT >

The possibility of the occurrence is only 11.4%, but combinations with these 4 blocks are the most difficult to deal with. Fortunately there is already a toolbox for them.  

ZSO: STD
ZST: pyramid
ZOT: pyramid
SOT: pyramid



Below listed is our solution 4 to play against combo:

TKI opening +
DT cannon opening +
TD cannon opening +
hanging opening(s) +
Pyramid opening +
zigzag opening


When a player is getting better, he/she can foresee and calcute with more pieces, thus will adjust the usage rate of these openings according to his/her personal preferrence.