HOME WORLD FORUMS WIKI VIDEOS
9580 members and stacking!
     Welcome guest, please login or sign up

5 Pages V < 1 2 3 4 5 >  
Reply to this topicStart new topic
> Hold that T piece! A parity experiment...
UJS3
post Feb 24 2013, 05:29 PM
Post #31


Tetris Apprentice
Group Icon
Posts: 147
Joined: 21-November 12



QUOTE(Paul676 @ Feb 24 2013, 04:57 PM) *

Also, my view is that since only placing a T or skimming which removes b2b bonus creates uneven parity, it must be the best players who know how to manipulate even parity fields for a tsd.

One other way to effectively create odd parity is to create holes in the stack. For example, the way an S/Z piece is placed in the corner in ST stacking effectively creates odd parity at the top surface without breaking B2B or using up a T. Of course, as you say, good players already use this all the time.
User is offlinePM
Go to the top of the page
+Quote Post
post Feb 24 2013, 07:13 PM
Post #32




Posts: 0
Joined: --



So in summary, pro japanese players are manipulating parity and creating odd parity on purpose to make easier TSDs. A very interesting concept. Theoretically, it all makes sense since odd parity tends to be jagged and thus suitable for a TSD hole.

As far as I know, a player can do one of two things that will change the parity:

1) Place a T in the field
2) Skim a piece, so that an uneven portion of the piece will be skimmed off

In both scenarios, the resulting field would have a reversed parity and thus an uneven balance of white and black. So, if these are the only possible ways to change parity, are these the only techniques we must practice to make TSDs at the pro japanese level? Or are there other ways a player can change his field parity, that I did not take into account?

And surely, simply having an odd parity field itself does not necessarily guarantee an obvious TSD. I'm sure the Japanese know something beyond just parity that enables them to consecutively make these TSDs. For instance, similar to how Blitz combined two T-pieces while stacking to preserve even parity. While it is true that combining two T-pieces theoretically preserves parity, he did not just combine them in any order, for there were some combinations that actually worsened the parity. Likewise, perhaps these japanese players are indeed creating odd parity, but only in a way in which the field would actually be benefited.


p.s. What do the pro japanese players do after they make a TSD and their field is now back to even parity? Paul676 suggested they repeat the process and make uneven parity again, but is this really the case? If, on the other hand, it is true that they are making TSDs even while in even parity, then it seems we are lacking of some further understanding.
User is offlinePM
Go to the top of the page
+Quote Post
myndzi
post Feb 25 2013, 01:21 AM
Post #33


Tetris Grand Master
Group Icon
Posts: 1,932
Joined: 26-June 09



QUOTE(jujube @ Feb 24 2013, 06:11 AM) *

Maybe one day it will be proven that always going for specific stack shapes at every piece is playable with 100% hard drops, and considered best if there's a method for doing it without much thinking.


I'd argue that Ryan Heise's AI is strong evidence for this being the case

QUOTE(Panda @ Feb 24 2013, 07:13 PM) *

Or are there other ways a player can change his field parity, that I did not take into account?


As UJS3 and I pointed out, the only parity that matters is along the stacking surface. Creating holes can change the parity too. In fact, I'm sure there are lots of ways it can be done, and the important thing isn't figuring out how -- it's being aware of it in your stacking.

An interesting thought: Since you can create holes to create a parity imbalance, then clear to reveal those holes, you can technically maintain b2b through consecutive TSDs with uneven parity. See, for example, the Albatross opening's TSD -> TST; uneven parity is created by the Z-prop that leaves a hole, and then when the hole is revealed by the TSD you've placed a T, leaving you once more with uneven parity.


--------------------
User is offlinePM
Go to the top of the page
+Quote Post
jujube
post Feb 25 2013, 01:28 AM
Post #34


Tetris Grand Master
Group Icon
Posts: 1,200
Joined: 26-June 09



QUOTE(Blitz @ Feb 24 2013, 04:25 PM) *

Until now, we have been pleased thinking japanese players are better because they are asian. I know this might come as a shock to you, but this is actually wrong. The reason why japanese players is in a bracket of their own, is because of their understanding of parity. Their understanding of parity is helping them to do more t spins. This might sound strange, but it is true. I think the reason this has not leaked out earlier, is because japanese people speak japanese, and japanese people understand japanese, so this knowledge passes on from one japanese to another, while everyone else doesn't understand a word of what they just said.

This is so true. I think there's also a conspiracy against online translators so that we'll never make sense of even the most basic japanese phrases.

QUOTE(myndzi @ Feb 25 2013, 01:10 AM) *

I'd argue that Ryan Heise's AI is strong evidence for this being the case

I'll have to check it out.

I wonder if players are intentionally making T-spin singles to change the parity for likely T-spin doubles. Go go Yoshihiro in any game that rewards EZ-TSS.
User is offlinePM
Go to the top of the page
+Quote Post
Paul676
post Feb 25 2013, 01:31 AM
Post #35


Tetris Grand Master
Group Icon
Posts: 2,491
Joined: 22-July 09



The more I think about parity, the more I think we all knew this anyway, whether subconscious or not, and this is a bit of a storm in a teacup.

EDIT:

http://www.youtube.com/watch?v=ipNc_IHNHI8

Just watched this to see if I could prove anything about parity, and how top Japanese players use it and whether Hebo_MAI's game evidences any advanced thought on parity. At first I thought I saw this, but when I actually studied each instance, they didn't convert even to uneven parity - they were just nice shapes for t-spin minis (upside-down L and J holes are awesome for t-spin minis, I observed). I think it's more to do with instinct and spotting shapes than knowing parity.

Worth noting that his 1st t-spin mini at 0:23 got him into bad territory because it gave him even parity from odd parity.

Unrelated to parity, this is absolutely lovely, if you care to learn that a 3-block gap between one tower and another is great for a laid down s/z air hook:



Another thing which is awesome if you wanna learn it is when to see this shape:



--------------------
User is offlinePM
Go to the top of the page
+Quote Post
myndzi
post Feb 25 2013, 01:44 AM
Post #36


Tetris Grand Master
Group Icon
Posts: 1,932
Joined: 26-June 09



QUOTE(Paul676 @ Feb 25 2013, 01:31 AM) *

The more I think about parity, the more I think we all knew this anyway, whether subconscious or not, and this is a bit of a storm in a teacup.


Well, it's like I was saying - this gave me the words to describe what I already knew. However, knowing the logic of the concept leads to greater understanding, which is good too.

As for keeping/tracking parity, I think it's much simpler than is implied here - you don't need to count skims and T-pieces etc, it really just comes down to the mino count on the surface. If an odd number of minos are exposed to the "air", it's odd; if even, it's even.

For perfect clearing there are two kinds of parity - there's the every-other-row thing since you can only PC rows in multiples of two - and there's the odd-block thing, which can be affected by garbage or stacking. I'm not sure the every-other-row thing has much use in general stacking but it is useful to pay attention to for perfect clearing.

QUOTE(jujube @ Feb 25 2013, 01:28 AM) *

I'll have to check it out.


Just imagine taking every field shape with a column-to-column difference of <= 4 and running them through PageRank - in the sense that every shape that can arrive at "this" shape is a vote for "this" shape being good. The most achievable shapes get the highest score.

The highest scoring shapes were in general flat, and in particular, the best scoring shape is:



--------------------
User is offlinePM
Go to the top of the page
+Quote Post
zaphod77
post Feb 25 2013, 04:32 AM
Post #37


Tetris Expert
Group Icon
Posts: 317
Joined: 31-July 09



QUOTE(Integration @ Feb 24 2013, 08:42 AM) *

The stack is still not fine. Assume your queue is J,O,O,S.



The problem is not that the stairs go up on the right side. This stack wouldn't be much better:




Aren't you forgetting about hold?

Lets say hold is, for some reason, empty. You can either hold that S, or hold the J, and work something out.

If it has an S in it, you would just pull it out first.

If hold also has a J, you can still hold that s.



also, clearing lines can only unbalance parity if it is unbalanced above and below the line clear after placing your piece, AND it's an ODD line clear.

A trivial example is to drop a S or Z vertically, and drop an I in very other column. once the single i cleared you will have it unbalances in one direction by two.

Even line clears NEVER affect parity in any way.

Besides global parity imbalance, local parity imbalance is possible. WHen the party below the line clear is imbalanced, AND the number of lines cleared is odd, you can create a global parity imbalance which will help you get more TSDs.

User is offlinePM
Go to the top of the page
+Quote Post
myndzi
post Feb 25 2013, 05:08 AM
Post #38


Tetris Grand Master
Group Icon
Posts: 1,932
Joined: 26-June 09



QUOTE(zaphod77 @ Feb 25 2013, 04:32 AM) *

Aren't you forgetting about hold?

Lets say hold is, for some reason, empty. You can either hold that S, or hold the J, and work something out.

If it has an S in it, you would just pull it out first.

If hold also has a J, you can still hold that s.
also, clearing lines can only unbalance parity if it is unbalanced above and below the line clear after placing your piece, AND it's an ODD line clear.

A trivial example is to drop a S or Z vertically, and drop an I in very other column. once the single i cleared you will have it unbalances in one direction by two.

Even line clears NEVER affect parity in any way.

Besides global parity imbalance, local parity imbalance is possible. WHen the party below the line clear is imbalanced, AND the number of lines cleared is odd, you can create a global parity imbalance which will help you get more TSDs.



In general, any sequence of pieces that you could manufacture with hold, you could also receive without hold. So for the purposes of discussing options, it's kind of a moot point -- if you could fix it with hold, you could have gotten that sequence naturally, and if you MUST fix it with hold, you can get a sequence that hold won't let you fix it with.

Also


--------------------
User is offlinePM
Go to the top of the page
+Quote Post
UJS3
post Feb 25 2013, 09:17 AM
Post #39


Tetris Apprentice
Group Icon
Posts: 147
Joined: 21-November 12



QUOTE(zaphod77 @ Feb 25 2013, 04:32 AM) *

Even line clears NEVER affect parity in any way.

This is true when clearing an even number of consecutive lines. A non-consecutive double is more like two singles though, minos change color and parity can change:



QUOTE(myndzi @ Feb 25 2013, 05:08 AM) *

In general, any sequence of pieces that you could manufacture with hold, you could also receive without hold.

Unless you hold across an entire bag, yes. As for your counterexample, zaphod is comparing the parity of the field right before the lines are cleared (after the line clearing piece is placed) to the one after the lines are cleared. From this point of view, placing the T changes the parity, while the line clear leaves it unchanged.
User is offlinePM
Go to the top of the page
+Quote Post
Blitz
post Feb 25 2013, 07:29 PM
Post #40


Tetris Specialist
Group Icon
Posts: 231
Joined: 20-April 12



QUOTE(Paul676 @ Feb 25 2013, 01:31 AM) *



http://www.youtube.com/watch?v=ipNc_IHNHI8

Just watched this to see if I could prove anything about parity, and how top Japanese players use it and whether Hebo_MAI's game evidences any advanced thought on parity.

I don't think that video of blink vs hebo shows any conscious thought on parity. According to my source (HAHAHA1744) the japanese players have only consciously known about parity for a year now. That video is older than that. But I still think blink and hebo were thinking about parity subconsciously.
If you are going to study games, you should watch a more recent video. I happened to record some games of subversive vs hahaha 2 days ago, so this is really fresh. I havent studied parity in theese games myself, but I'm pretty sure both hahaha and subversive plays better now, than what blink and hebo did back when their match was recorded.

This is bad quality cause I can't achieve better without crashing my pc.

User is offlinePM
Go to the top of the page
+Quote Post
zaphod77
post Feb 25 2013, 08:01 PM
Post #41


Tetris Expert
Group Icon
Posts: 317
Joined: 31-July 09



I'd like to see this sequence that cannot be fixed with hold under bag. Smile.png

I simply refuse to believe there are piece sequences that you can be forced into with a modern randomizer with hold.

If there's a piece other than the L O O S in hold you can use it instead of any of the pieces. If there's one of those three pieces in hold you can use it at a better time and fix the stack.

I'd like to see a sequence of 5 pieces, with the first one assumed to be in hold, that obeys bag and will force a hole. (the first piece is outside of the bags, because it was held earlier)

Only the need to eventually place a given tetromino can be forced even with hold, not an entire sequence.

And, well, a hurdle double is two singles. Smile.png It's true that a hurdle can change parity.

But the other important insight is that the parity of the entire field doesn't always matter. as i said dropping a z vertically onto flat ground creates two instances of local odd parity (one above the center line of the S/Z and one below it. placing holes in the stack deliberately can quickly generate a local parity imbalance that can be used to form a t-slot. And once the tsd is made, you then are left with another parity imbalance afterwards.

Garbage, for the most part, doesn't affect parity meaningfully until you dig down to it (but then you can use it as a starting place for a t-slot.)



User is offlinePM
Go to the top of the page
+Quote Post
Paul676
post Feb 26 2013, 01:17 AM
Post #42


Tetris Grand Master
Group Icon
Posts: 2,491
Joined: 22-July 09



Sorry Blitz, I could see no conscious thought on parity here. But I'm not as experienced as you at it so maybe you could point it out to me?


--------------------
User is offlinePM
Go to the top of the page
+Quote Post
post Feb 26 2013, 02:15 PM
Post #43




Posts: 0
Joined: --



QUOTE(Paul676 @ Feb 26 2013, 01:17 AM) *

Sorry Blitz, I could see no conscious thought on parity here. But I'm not as experienced as you at it so maybe you could point it out to me?


I gotta admit, many of us were probably over-excited by the fact that a single concept (parity) could explain the reason for Japanese dominance. It was too good to be true, and indeed some reflection and sanity convinced me that not even the Japanese can consciously calculate and track parity in an actual live match. Sure, they may take parity into account when it is obvious or instantly detectable, similar to how an amateur would make a t-spin if it is conveniently there, but it is certainly not a continuous procedure, not at the speed and pressure they play with.

All in all, parity was an interesting topic with theoretical truth and importance behind it. But it is not the reason for Japanese dominance. At least not by itself.
User is offlinePM
Go to the top of the page
+Quote Post
zaphod77
post Feb 26 2013, 04:50 PM
Post #44


Tetris Expert
Group Icon
Posts: 317
Joined: 31-July 09



Well, kit's not the parity of the entire field that you need to keep track of, it's the parity of the area you are working with that's important.

vertically placed S and Z are the simple way to create t-notches without using a t. even though the global parity doesn't change, if you divide the field into the part containing the upper part of the snake, the part containing the middle, and the part containing the lower part, you will fid there's a parity shift in two out of the three zones, whihc allow syou to restore that sction parity later by spinning in a t.

TO get more TSDs, you have to learn to set up t-slots (local parity imbalance) without using the Ts. Is it theoretically possible to gt a TSD with every t piece received?
User is offlinePM
Go to the top of the page
+Quote Post
post Feb 26 2013, 06:13 PM
Post #45




Posts: 0
Joined: --



QUOTE(zaphod77 @ Feb 26 2013, 04:50 PM) *

Well, kit's not the parity of the entire field that you need to keep track of, it's the parity of the area you are working with that's important.

vertically placed S and Z are the simple way to create t-notches without using a t. even though the global parity doesn't change, if you divide the field into the part containing the upper part of the snake, the part containing the middle, and the part containing the lower part, you will fid there's a parity shift in two out of the three zones, whihc allow syou to restore that sction parity later by spinning in a t.

TO get more TSDs, you have to learn to set up t-slots (local parity imbalance) without using the Ts. Is it theiretically possible to gt a TSD with every t piece received?


Local parity imbalance? That's an absurd topic, that's even harder to calculate than the overall field parity and defeats the purpose. At that point, it is no longer about parity but simply shape detection and adaptation, as you mention with the S and Z setup. Local parity calculation is incredibly complex and is unrealistic to perform in a live match. Impossible even.
User is offlinePM
Go to the top of the page
+Quote Post

5 Pages V < 1 2 3 4 5 >
Reply to this topicStart new topic
1 User(s) are reading this topic (1 Guests and 0 Anonymous Users)
0 Members:

 

©2009-2014 Hard Drop Community & Forum
harddrop.com is not sponsored or endorsed by The Tetris Company or its subsidiaries.