What happened?

[Sisu]

Tetris from the Top: An Interview with Henk Rogers

"the next generation, the way I'm looking at the Tetris machine, we call it the 'Tetreon', the way I look at the Tetreon is it's kind of like a vehicle. So some things have to be standardized, like the brake and the gas pedal, but some things can be optional, like power windows and power brakes."

"the ability for a player to play versus mode against somebody else who's got their own machine that's built to a different spec. It's like instead of having Formula One where everybody drives the same kind of car, it's more like Le Mans where everybody gets to choose what kind of car they're going to drive."

"if we're playing head to head, and you have your favorite way of playing Tetris and I have my favorite way of playing Tetris, so you'll by a Tetreon which has the feature set that you like, and I'll buy a Tetreon which has the feature set I like. And I might buy some options to adjust it, to make it play exactly the way I like. Maybe I don't like how long it takes the piece to lock down, maybe I don't like the drop speed. All those things are eventually adjustable...and we could have a competition."

Henk nails it with this concept. He doesn't really flesh it out though. Tetris is missing a class-based system.

He gave it a shot with:

Super-kewl I know, but I believe the actual Tetris mechanics are the same for all characters. Gotta love the insta-4w spell at 4:10


Think more like:

[attachmentid=224]

[attachmentid=225]

[attachmentid=226]

[attachmentid=227]

If you want to go fast, fine. You're an elf

You might have to give up attacking and defensive abilities in exchange for no speed restrictions. On the other extreme you might have a "Zangief" who plays NES or Atari style but with the potential to cancel a 20-combo or send 10 lines with a single Tetris. So on and so on for different classes.

Larger unified player-base. More interesting match-ups. Yay.

[myndzi]

Let's do something about it.

[caffeine]

Everyone seems to equate that playing fast means playing good, which isn't the case. They don't realize that to be good you have to be efficient as well.
I think I understand what you are trying to say, but I don't think the game should be slowed down.

This is accurate. Slower, but smarter, players have been taking down faster players for ages. If you're not taking down faster players, it's because you're either not playing fast enough or smart enough.

Anonymous: I agree with what you say about players not knowing how to use their speed, but I would add that the biggest reason is because they don't have to. Playing faster gets many people by because the truth is that speed can compensate for less effective play. Stacking will tell between two players of the same speed, but there is a point at which poor play at high speed will triumph over good play at low speed, and it's attainable for many people, extreme cases excluded.

Speed can compensate for less effective play, yes. I'm going to copypasta something a portion of something I've been working on but haven't posted about yet:

Throughput: the total garbage a player adds to his opponent and removes from his own field divided by the number of tetrominoes he required to either perform an attack (e.g. the throughput of a single Tetris) or to win or lose the game (e.g. how efficiently they played throughout the entire game).

Flux: the total garbage a player adds to his opponent and removes from his own field divided by the number of minutes used to win or lose the game. Flux represents the rate of a player's throughput. For example, Player A might have a better throughput than Player B. Yet if Player B is fast enough, he can still outmatch player A. This is because even though he's less efficient, he's able to push that lesser throughput out fast enough to make up the difference. A player who can both play fast and generate a high throughput will produce the highest flux overall and will be a force to reckon with.

So, there's a simple relationship between raw speed, throughput, and flux: (TPM)*(throughput)=(flux). Of course, the variables that influence the outcome of real games make things much more complex than this, but this is the basic concept we're talking about. What it comes down to is that a player who is twice as efficient (aka has twice the throughput, aka sends and clears twice as much garbage per piece) will break even against an opponent who plays twice as fast. In terms of flux, speed and throughput are weighted equally. You can compensate one with the other, but it's best to maximize both. This is why trance can beat maserati (his throughput trumps mase's higher speed), and mase beats me (his speed trumps my higher throughput). (Note: if trance doesn't actually beat mase, sorry! That's just the impression i got from HDO3.) The beautiful thing about this relationship is that, often enough, playing faster means sacrificing throughput. Thus, you don't really play better by playing faster... that is, until you learn to play better... at a faster rate.

I understand what you mean about enjoying play with equal players at high speed, but the main thrust of my comment is that there aren't enough players for that. I'm not suggesting that high speed play be completely eliminated, but I do think that if a more moderate game (perhaps moderated in other areas too, such as garbage) was the norm instead of the exception, it would be more welcoming to a wider range of players. (As well as good sportsmanship/attitude!)

In the end, what has happened is that the skilled players (such as yourself, for example) have come to prefer settings that isolate them from the majority of players.

1) I think that Tetris is becoming more and more garbage heavy. We've gone from sending 4, 2, or 1 lines in very restricted circumstances to being able to send large amounts of lines quickly with combos, allspins, perfect clears, and Change on Attack garbage.
2) I think that Tetris players have an unhealthy love affair with speed. They want low DAS, instant AR, firm drop, hard drop, and so on.
3) Both of these factors contribute significantly to accelerating the pace of the game.

It is my belief that, in a slow enough game, Bar can have an enjoyable game against Foo. It will be difficult and he won't win many, but he won't get dominated so hard that he's not even interested in playing.

In a fast enough game, on the other hand, Foo will dominate Bar, and Bar won't want to play unless he's a masochist (and I grant that many of us are indeed the kind of players who might play 100 games and never win a one of them).

Furthermore, I believe this effect compounds at higher ratings. If Foo and Bar have an average rating of 2000 and the "range" within which they can have enjoyable matches is +-100, I believe that range shrinks if their average rating is higher. If their average rating is 3000, maybe that range is +-20. This is most obvious in large tournaments where the third place players are lucky to take one game in two, or even one in three, against the winners.

So what I'm trying to say is that by slowing down the pace of the game, we can widen the range of enjoyable matches. You might not win any more frequently, but you also won't lose as hard, or at least, you'll have a better chance and it won't be over before you can even get into the game. I think all the extra garbage adders could do to be toned down for similar reasons.

So, you're making the argument that to undermine the influence of TPM in the equation (TPM)*(throughput)=(flux), in that the slower player's TPM will represent a larger percentage of the faster player's TPM since both player's TPM will be negatively affected, this will make the game more enjoyable for at least one of the two players (the slower player). You're claiming that this will increase the userbase. I see a couple of problems with this idea.

1. Decreasing the potential speed of the game also decreases the overall quality of the game. Responsiveness is a good thing, and players enjoy pushing their limits. Messing with DAS also means messing with finesse schemes that some players have put a great deal of time into learning and optimizing.

2. The userbase you're trying to access already can't compete with faster players, whether or not they're brought down to equal speed. (With added delays, faster players will still be faster, but let's pretend speed will equalize). The reason they can't compete is because they're not only slower but also less efficient. This is why I can join a speed-limited game in Cultris 2, carefully watch my speedometer, and always win against the noobs. I can downstack and they can't, pretty much. It has nothing to do with speed. In fact, the only player-base that lowering the speed threshold would affect is the players who have a higher throughput average but not enough speed to beat the speedfreaks who dominate them. This is a subset of a subset of players that are already "in for the long run." I'd say they're actually harder core than the average hardcorer. My opinion, then, is that this will not allow players to be dominated any less than they already are, other than the fact that these unevenly matched games will last a little longer. I believe this "extra time 'till death" is what you are advocating will make the game more enjoyable for the weaker player. However, all this extra time is made up of waiting frames and not the enjoyable frames where the player gets to do fun stuff. To risk oversimplifying, it's like saying Bar will get an extra second of gameplay when he plays Foo, but that extra second is composed of a one second line clear delay before he tops out. My opinion is that you're not really adding any quality to Bar's experience by doing this.

I wouldn't want a game mode where I could beat Blink in a first to fifteen. I wouldn't want one where I could beat him first to five either. Those would both be stupid. I'd like a game where I could at least feel like I could beat him on occasion. It's not about shuffling the rankings of the best players, it's just about condensing them together a bit so that we get far more entertaining and closely-matched games.

For what it's worth, I personally think that Tetris DS (both 2p and 4p) had it pretty much nailed for everything regarding multiplayer - both close and entertaining matches, and a decent rating system. The interface for match-making wasn't the best by a long chalk, but as far as competitive gameplay for any official game has gone, it was easily the best I've played.

Exactly what change are you saying would make you feel like you could have a chance to beat blink on occasion? Slowing down the game? If that's the case, do you think you'd have a chance at beating blink in a zero-speed game such as KOS? Or, do you mean by adding a luck factor which will allow you to get an advantage by means other than your skill? If that's the case, do you really think that winning due to lucky circumstances will feel as representative of your effort and feel as rewarding as winning without lucky circumstances? In any case, I agree with you that the combination of matchmaking and large playerbase made TDS fun.


MEGAPOOOOOOOOOOOSTT!!!!!1!

[myndzi]

Wow, that's a really thoughtful analysis caffeine. I'd like to separate two of my statements that I may have gotten mixed up. I don't know believe that a slower game will directly increase the userbase. I am only suggesting that it will make a larger range of matches enjoyable than otherwise.

I think that it will do this in two ways.

1) It will allow players with a greater speed difference to play matches that they couldn't otherwise. Imposing overhead delays will necessarily bring the two players' speeds closer together. This will be true at any skill range.

2) It will draw out the game, which grants a greater impression of struggle than just getting topped out right away. It might be inevitable, but it seems like more of a fight.

Your description of throughput and flux makes sense, but I think it risks being an oversimplification on some important parts, not the least of which is thinking time. By taking the time-pressure down a notch, it allows a wider group of players to do more with the game than just struggle not to die right away.

Regarding this talk of "subsets", let's divide things up.

1. Equal matches - which matches are equal may vary based on the balance of flux versus speed if you go about altering one of the variables, but I think there will be about the same number of equal matches unless you go way overboard with the changes.

2. Unequal but close matches - these matches have at least a chance of success for the underdog. Say 1:2; the winner isn't likely to be in question in a best of 15 set, but the better player can't let their guard down and the lesser player can take matches with exceptional play.

3. Unequal and not close matches - these matches are pretty much hopeless. The winner may get bored of beating the snot out of the loser, and the loser may get depressed about not having a chance.

The idea, for me, is to take some category 3 matches and make them category 2 matches. That is, I'm not trying to make it more likely that less skilled players will defeat more skilled players, but I would like it not to  be a total loss. This is important because we don't have a large community, so it can be hard to find appropriately matched opponents to have evenly matched games with.

Now, let's consider how compressing the speed by imposing various delays affects various matchups.

Let's use letters now! Let's say type A players are players who have high speed and low throughput. Type B are approximately equal. Type C are high throughput low speed.

Group 1: B vs B will go mostly unchanged. A vs B will benefit B. C vs B will benefit C. A vs C will benefiit C by a lot. A vs C matches may wind up in group 2 after such a change, depending on its degree. The others are probably still going to remain in group 1.

Group 2: A winning against B will become a closer match, possibly enough to move it into group 1. B over A will get worse, but probably not enough to move down. A over C will become a much closer match, maybe moving up. C over A will probably move down. C over B will get a little worse and B over C will get a little better, both probably stay.

Group 3: A over B: probably moves up. A over C: probably moves up. B over C: probably moves up. B over A: probably stays. C over A: stays for certain. C over B: probably stays.

In summary, high throughput:speed players benefit, high speed:throughput players are hurt. Some (~1/6?) matches from group 1 will move to group 2; some (~1/6) will move from 2 to 3. About half will move from 3 to 2.

If you agree with my admittedly hasty categorizations, I think this shows that reducing the impact of speed on the equation will bring more games out of group C, which I think is desirable. It also pretty much universally benefits balanced players, which I think is desirable.

So, while your original statement about my comments "addressing a subset of a subset" may be reasonable, I believe that not every portion of everything needs addressing. I think that addressing a small portion of poor games in a way that can make more of them better is beneficial, particularly when the option of avoiding such matchups is not necessarily available.

vvv that's part of why I qualified some of the "stays maybe" or "moves up/down maybe" categorizations. For one, with A vs B or B vs C, the speed difference isn't as great, and so compressing the speeds doesn't change the relationship as much as A vs C. For the other, more speed = more idle time = more time to think. Some players, like meow, get ridiculously scary in these situations, so it's certainly the case that even players with medium speed and high throughput can increase their throughput even more, but it's not guaranteed. Really it depends on the player, which is why I don't think you can say that anything will happen universally; no matter what two players' comparative speeds are, they could both increase throughput if forced to slow down.

[Paul676]

I'd suggest that for high speed low throughput players, when speed is artificially lowered, their throughput becomes higher, and are not necessarily hurt. This challenges your assumption that this isn't the case, meaning that arguably (anon has already made this point anyway) the result won't change much.

As for throughput/flux, it's a great analysis, and I hope that somewhere along the way the "spike factor" i.e. how irregular player x's attacks are, is put in

[caffeine]

First off, thank you for taking the time to extend the people skills that are so rarely seen among forum debates. I can tell that even though we disagree on how to go about getting the desired result, you're in this for the right reasons, and I hope you find some success with all of this.

Now, let's consider how compressing the speed by imposing various delays affects various matchups.

Let's use letters now! Let's say type A players are players who have high speed and low throughput. Type B are approximately equal. Type C are high throughput low speed.

Group 1: B vs B will go mostly unchanged. A vs B will benefit B. C vs B will benefit C. A vs C will benefiit C by a lot. A vs C matches may wind up in group 2 after such a change, depending on its degree. The others are probably still going to remain in group 1.

Group 2: A winning against B will become a closer match, possibly enough to move it into group 1. B over A will get worse, but probably not enough to move down. A over C will become a much closer match, maybe moving up. C over A will probably move down. C over B will get a little worse and B over C will get a little better, both probably stay.

Group 3: A over B: probably moves up. A over C: probably moves up. B over C: probably moves up. B over A: probably stays. C over A: stays for certain. C over B: probably stays.

In summary, high throughput:speed players benefit, high speed:throughput players are hurt. Some (~1/6?) matches from group 1 will move to group 2; some (~1/6) will move from 2 to 3. About half will move from 3 to 2.

If you agree with my admittedly hasty categorizations, I think this shows that reducing the impact of speed on the equation will bring more games out of group C, which I think is desirable. It also pretty much universally benefits balanced players, which I think is desirable.

So, while your original statement about my comments "addressing a subset of a subset" may be reasonable, I believe that not every portion of everything needs addressing. I think that addressing a small portion of poor games in a way that can make more of them better is beneficial, particularly when the option of avoiding such matchups is not necessarily available.

I won't go into detail with your first two points, as I believe I've already addressed them. Good job on the subsets, as this all seems to be a very good assessment of what lowering the speed threshold would do to the matchups. The only thing is that I believe it's only true for cases where flux is the same in the first place (Group 1). Group 3 is the one we're worried about, but as I elaborated on before, this is the group where the A players' throughput is still better than the weaker players in the first place, so they'll still dominate.

As for throughput/flux, it's a great analysis, and I hope that somewhere along the way the "spike factor" i.e. how irregular player x's attacks are, is put in

Thanks. I have this big document that goes over many different aspects of my flux model, and I also have quite a few excel sheets that analyze how this model quantifies different in-game situations, which I believe can reveal a lot about how we can best play the game. I'll get around to posting it all one day, but it's just not ready yet. That's an interesting thought about the "spike factor." I'm trying to account for as many variables as possible, but at some point you just have to make it "Tetris in a vacuum" to be able quantify things. However, I will say that as your field becomes higher relative to your opponents, clearing garbage becomes more valuable than sending garbage (and vice versa in the reverse situation). Another way to put it is that clearing down garbage is meaningless when a quick TSD would top your opponent out.

[myndzi]

Well, maybe there needs to be a group four: Players that are completely outclassed. In my response, I was only considering cases we can do something about. Of course there are going to be cases that are "I've only been playing Tetris for 3 days" - and there's no way they'll ever have a good game against people who've played for 9 years without extreme handicapping going on. Not even then, really. I have played 1 versus 5 with a 16 line handicap in Tetrinet and still pulled it off regularly, and I know there were plenty of players better than me.

So yes, I guess I implicitly (and unthinkingly) performed my analysis based on the set of players within a certain range, but I posit that no matter what you do, the other cases can't be helped regardless. If we take the cases we can do something about, which is to say, when matching up players in my "A, B, C" sets, no player outranks the other player in both attributes, do you accept the point as valid? (And even if an A player has better throughput than a C player, it's likely that nerfing the speed will bring them closer together and therefore improve the game slightly even if the A player will still dominate.)

I guess the question then is how many players are really so far out of range that nothing can help? Or better, how many players are close enough that a little boost could improve their experience?

I also propose that in general the slower and less skilled the players are, the greater the range is before the match is pushed into group four, and that this effect is widened by slower games.

[mippo]

Uhhh, I don't have a large study I'm working on that I can use in my argument. I feel I'm kind of fast but not ridiculously fast.
I'm still beatable by people who are a lot slower than me (I guess around 10 secs?) and I sometimes beat people faster than me.
I very rarely really have the feeling that people beat me because they are faster, just that they spiked better or played more efficient.

So any idea on the range of TPM ranges that you are thinking of?
I know Meow can beat me going 0.5 - 0.7 pps slower.

Instead of focusing and limiting speed, I'd rather have a slightly more random bag (or even no hold button). This would slow down play, have people be more creative and also nerf combos.
I don't think it should be totally random just slightly more random than the 7-bag (or whatever it's called) that is used in most games. I don't know what would make a good balance, you should still be more or less able to rely on pieces, maybe something like the history 4-bag or a lager bag of with random pieces added to each?

[Paradox]

[!--quoteo--][div class=\\\'quotetop\\\']QUOTE[/div][div class=\\\'quotemain\\\'][!--quotec--] The point is that unlimited speed restrictions lead to less of a difference in player skill making an exponentially greater difference in ability to compete.
[/quote]

I'm saying its not speed to blame for the skill difference, its just overall skill. People who are more committed win and those who are not lose.  I'm trying to point out that you shouldn't blame speed. Good players are fast, its just part of the set of skills they have. That's like blaming the problem on t-spins because some people are much better at it than others.. but it just  means they have practiced well. Same with speed?

The part about me saying "you should play even if you are slower because it takes time " was just addressing how you said you don't play multiplayer very much because you aren't as fast.

[!--quoteo--][div class=\\\'quotetop\\\']QUOTE[/div][div class=\\\'quotemain\\\'][!--quotec--]it's not fun to play when the only way to have a chance is to invest more time than I am able or willing to invest in getting faster[/quote]

Also I'd like to point out that speed is not the biggest factor in skill. People tend to think that speed is overpowering if they lack in speed. I know plenty players who are really good and are not that fast. I just don't want the blame to be put on speed and will think they have to add delay to make the game fair.

With the current difference in skill levels I don't think there is anything really to blame. I think its just the nature of the game. In tetris people are rarely at the same skill level- (maybe its because competitively we are few in number?)

And sc2 rocks, yes the multilplayer there is very mature and its just a blast to play. Te

[myndzi]

Since you obviously didn't try to understand what I was getting at as well as some other participants in this thread did, I'm not going to continue this argument with you until you have done so

(Hint: read caffeine's post)

[Paul676]

a lager bag of with random pieces added to each?

mmm lager - drunk bag forgets its own pieces

[thelulz]

I wanna tell you 'bout a newbie that's been plaguing the forums
He's like the the veteranest newbie, but he acts like the quorum
In charge of handing out l33tness like it's kept in a basket
But he shares it with noone, not even if you ask it

Now I hear he once was better than the average player
He gained a reputation but I wouldn't place a wager
Whether hither or yon, I've only ever seen him quit
In a flurry of curses after getting a bit
Defeated


Co-co, Corrosive
He'll tell you that he's great until he's blue in the face
You know I've never really seen him win a game, though
Co-co, Corrosive
He thinks he was an ace, that he'll put you in your place
You know I've never really seen him earn his fame, though


When he posts in a thread, you know there's nothing to read
It's all in his head, and though he's preaching the creed
I don't know any converts, I never did see
Anybody put effort into "getting like me"

I'll give him some credit, he can play pretty fast
But it's counteracted by the way he talks out his a**
And maybe just a little bit by how he never stacks clean
But I shouldn't make fun, 'cause it really is mean


To talk a big pile of sh** about his sh**ty a** stacking
When there's better things to rip on; I don't mean to send him packing
But his natur'l inclination is to cry and run his mouth
And so before he even gets there I'mma get the ʞɔnɟ out


Co-co, Corrosive
He'll tell you that he's great until he's blue in the face
You know I've never really seen him win a game, though
Co-co, Corrosive
He thinks he was an ace, that he'll put you in your place
You know I've never really seen him earn his fame, though

I've never really seen him win a game, yo

[Sisu]

I wanna tell you 'bout a newbie that's been plaguing the forums
He's like the the veteranest newbie, but he acts like the quorum
In charge of handing out l33tness like it's kept in a basket
But he shares it with noone, not even if you ask it

Now I hear he once was better than the average player
He gained a reputation but I wouldn't place a wager
Whether hither or yon, I've only ever seen him quit
In a flurry of curses after getting a bit
Defeated
Co-co, Corrosive
He'll tell you that he's great until he's blue in the face
You know I've never really seen him win a game, though
Co-co, Corrosive
He thinks he was an ace, that he'll put you in your place
You know I've never really seen him earn his fame, though
When he posts in a thread, you know there's nothing to read
It's all in his head, and though he's preaching the creed
I don't know any converts, I never did see
Anybody put effort into "getting like me"

I'll give him some credit, he can play pretty fast
But it's counteracted by the way he talks out his a**
And maybe just a little bit by how he never stacks clean
But I shouldn't make fun, 'cause it really is mean
To talk a big pile of sh** about his sh**ty a** stacking
When there's better things to rip on; I don't mean to send him packing
But his natur'l inclination is to cry and run his mouth
And so before he even gets there I'mma get the ʞɔnɟ out
Co-co, Corrosive
He'll tell you that he's great until he's blue in the face
You know I've never really seen him win a game, though
Co-co, Corrosive
He thinks he was an ace, that he'll put you in your place
You know I've never really seen him earn his fame, though

I've never really seen him win a game, yo

HDO I

js

[CorrosiveX]

I see this thread I created failed to bring about any change so far. People starting a few tournaments and contests and "earn a tetris belt color" isn't going to establish any unity for us all to be on one damn game. Multiplayer Tetris didn't birth on a FORUM WEBSITE, it started on a GAME. The Tetris scene that exists today is pathetic and is turning into TGM, where people only play and then talk about it on a forum. Nullpomino is like a drive-through or a Walgreens where people drop in for a few seconds and leave.

Whatever old school TNET2 and tfast servers had going on back in the days before all this t-spin and combo BS, they had it right. There was unity and a lot of people playing the game at the same time. There was rooms full of people and there was there was a lot of competitiveness and fun without any need of tournaments or ladders. Few of you have even experienced how fun it is to play a 3v3 or 2v2v2. There have been no drastic actions taken to try to unify everyone onto one game, and nobody cares. If you all just want to chat on the shoutbox then we might as well all be on a site like Habbo Hotel and creating little avatar people and buying furniture since you don't want to play Tetris against each other. Every clone has a chat system which would support 50+ people to constantly log onto and communicate with each other. For the people who claim to run things around here, Kill the shoutbox and cut ties with Tetrisfriends.com, and start figuring out how to unite people onto one golden clone like it's supposed to be.

p.s. KING OF STACKERS??? I want to puke. You guys throw TOP out there and now this. Just another game to divide everyone up even more. KoS is not an original idea by any means, tnet1 was practically turn-based by the incredible delay time that it had inbetween drops. If you want to master your stacking then you could have just had people go play tnet1, or any game where people simply TAKE TURNS DROPPING.

SUPPORT MULTIPLAYER UNITY