Blockbox

More like we've all got so much to learn from blink. Am I right?

 

Heh, very true

 

Is Blockbox and Blockendoos the same?

 

Yes.

 

If there are some more good players out there I might break my vow to try and give up keyboard Tetris for lent to play some Blokkendoos (acceptable, because I only made the vow to try and play Texmaster exclusively with my stick). I feel bad joining games where the players are closely-matched beginners and just annihilating them, but it's hard to find other games.

Also, if it's not already been done, I'd ask that the rating system please be adjusted/fixed for games involving more than two players. I think the current way it calculates increases and decreases in rating is really irritating and it especially doesn't reflect how multiplayer games actually work.

 

You know where it says your grades, TPM, and time played? Instead of straight average TPM, what about a rolling TPM? Right now it's very misleading because of single player modes. For example, it says blink's TPM is under 105, yet we see him consistently finishing games in the 170s. =b By rolling TPM, I mean you could have the average of the last twenty games played. This should be pretty accurate. The added benefit is that if a player starts playing blockbox and sucks big time, his early days won't have a permanent negative effect on his TPM.

What do you recommend?

 

Yeah, that was bothering me a bit as well. Not because it underestimates my ability (that can be a great surprise for the opponent!) but because the stat becomes irrelevant. Oh, and a rolling TPM would also more accurately reflect a drop in TPM if your ability took a dip for whatever reason.

 

Currently it works thusly: rating change = (change vs average of players beaten) - (change vs average of players beaten by).

This is fine for two-player, but I just don't think it works at all for higher numbers of players. If I'm playing against a 2000-rated player, and then a 1000 player joins, I'll get significantly less points per win, despite the games actually being at least the same difficulty. I still have to beat the 2000 player and now the 1000 player as well, and yet I get less points, and I just don't see how that really justifies itself.

It'd be fixed quite easily if the system didn't take an average anywhere, and just calculated the rating change against each player and then added up all the points gained for the wins minus the points lost for the losses.
I just think it makes sense that if you're playing more people you should get more points. Certainly not less.

 

Just get Joytokey. It works fine for using joysticks with Blokkendoos.

 

Yeah, but I'm never going to use a joystick for 0G versus Tetris with hard-drop enabled.

 

If you can't play low-G world rule on a stick, then try using the controller on the left:

(ignore the DS)

 

Less can't, more why would I when a keyboard is massively faster.

Couldn't you just use the D-Pad on the N64? That'd make it pretty much the same as the DS.

 

To make the game fairer, all players should receive the same string of garbage. For example, if Player A's first garbage row has a hole in column 2 and his next in column 6, so should player B's. I drew a picture to illustrate how uneven garbage can be sometimes.



One player gets really evil garbage while the next player gets very easy garbage. Player B will have a large advantage in this game.

 

Yeah, as caff pointed out, it's not just consecutive hole alignment that is the problem. Even if the hole was to change every time, something like Player A's garbage pattern can still occur as long as the possibility is there.

You can go to extremes and design a complicated garbage system to eliminate this type of pattern. However, that's just diluting the problem and not really solving it. If that disadvantageous pattern is eliminated, a new one will take its place as the most disadvantageous pattern. Plus, you would end up reducing the possibilities of the garbage to a really small (possibly arbitrary) pool.

You can even apply a randomizer system like 7-bag or history to this smart garbage system. However, the unfairness is still present in the short/long term. The most elegant and easiest solution is to just have both players get the same garbage pattern.

 

Which shouldn't be difficult in an object-oriented system. You're using Java, not something restricted like assembly or C on a sub-20-MHz CPU. So you can make your garbage generators implement something like this, where each player gets the same seed value:

Code:

interface GarbageGenerator(w) {
 /**
 * Reinitializes the generator.
 * @param width the width of the playfield
 * @param difficulty a number from 1 to 100, interpreted
 * @param seed a number identifying which sequence is produced
 */
 void reset(int width, int difficulty, int seed);

 /**
 * Makes the pattern for one row of garbage.
 *
 * Each bit represents one cell (1 for filled, 0 for hole).
 * The least significant bit represents the rightmost cell.
 * @return bitfield of which cells are filled
 */
 int makeRow();
}

(Because this interface uses a bitfield, not a hole X position, it's flexible enough to represent the TGM+ pattern, which has multiple holes per row, or even the Shirase pattern, which copies the playfield's bottom row.)

But if you give both players the same seed, players can send garbage, screen-peek, and then prepare their stacks accordingly. I used to do this all the time on Tetris in Tetris & Dr. Mario.

 

what if both players received the same randomness but in different patterns? the garbage hole changes would be in sync but the holes would appear random columns. it would only allow for a small advantage from screen peeking, but one could argue the player who strikes first deserves that insight.

i suppose those 1 column shifts in garbage (Player A in caffeine's post) could be copied to the other player's field, but a new hole could appear either 1 column left or right of the previous hole, instead of following the exact pattern the other player received. example:

 

But the trilling thing that caffeine described works entirely on position of the holes to generate unfairness, rather than how often the hole changes.

I'd love to see someone actually take the time to get an advantage from looking at their opponents garbage during a fast competitive match. I think at the sorts of levels where people would think to take an advantage like that, the gameplay is too fast for it to be worth the delay you require to shift your attention. Dr Mario is a damn slow game as it is, let alone when compared to 150+ tpm Tetris.

 

will there be another version of blockbox?

 

Yeah, of course. Deniax has some new modes up his sleeve (in testing). And I'm sure he has a long todo list he's working on.