Lowest KPT challenge

Started by belzebub, February 06, 2012, 06:15:36 AM

Previous topic - Next topic

myndzi

#15
Oh trust me, I almost went below 2.0 with some silly tricks - I just can't play fast enough to do them in under a minute and a half. I've probably thought of the same ones, I was hoping you had something clever I didn't think of

Edit: oh! I just thought of something good

Edit2: haha broke 2 KPT

bigwig

It was working when I did what you said, but now for some reason 0 gametypes show up and I can't play

Saturday

how do you get the Finesse to show ?

is there a setting i'm not aware of? I see LINE, PIECE, LINE/MIN, PIECE/SEC, TIME

Paradox

you need belzebubs nullpomino mod

if you want my copy of nullpo you can try it

http://www.mediafire.com/?t70z4ordlt9f7q0
[!--ImageUrlBegin--][a href=\\\"http://oi46.tinypic.com/2zqx63k.jpg\\\" target=\\\"_new\\\"][!--ImageUrlEBegin--][img width=\\\"400\\\" class=\\\"attach\\\" src=\\\"http://oi46.tinypic.com/2zqx63k.jpg\\\" border=\\\'0\\\' alt=\\\"IPB Image\\\" /][!--ImageUrlEnd--][/a][!--ImageUrlEEnd--]

xenoslash

Just a thought.
Maybe set the das delay to be 20 ish
At the start of the game press left, and never let go.
If need pieces to go right, just hold right, and let go if you reach the place.

Haven't tried this yet though.
"I just know everything." - kwillin


Kitaru

#21
@xenoslash: I'm pretty sure LapSi's rule ignores DAS Delay -- the iteration rate is always Instant. Also, if you were going to use skillstop finesse with such a large overhead, you should definitely never be letting go of left or right.

@caffeine: Nice.
<a href=http://backloggery.com/kitaru><img src="http://backloggery.com/kitaru/sig.gif" border='0' alt="My Backloggery" /></a>

xenoslash

#22
Quote from: Kitaru
@xenoslash: I'm pretty sure LapSi's rule ignores DAS Delay -- the iteration rate is always Instant. Also, if you were going to use skillstop finesse with such a large overhead, you should definitely never be letting go of left or right.

@caffeine: Nice.

Nah, caffeine used das delay too.
But if one is really good at spotting the right time to drop, movement is guaranteed to only taken one key press, 0 if its somewhere on the left side.
So choosing a setup which requires a lot of heavy building on one side, but continuous stream of pieces on the right will minimize the number of required movements.
I think that's why caffeine chose to do that 4 wide..

EDIT:
Come to think of it, it might be possible to finish the run with just <10 keys wasted on movement.
height of a 4 wide stack = h;
40/h = number of wides you need to do to complete the race

so 1 movement for left at the start of the game
for every 4wide, you just need 1movement to the right.
To resetup 4 wide, just release the right key.

total key spent on movement = 1+ 40/h

"I just know everything." - kwillin

caffeine

#23
Before everyone goes off saying we should change the rules (which is a good idea), keep in mind that even if you say "0G+instant DAS," you still have the hold bug to deal with. Even if you say "no exploiting hold," people that happen to use hold by chance would still benefit from it. So, my point is that adding rules won't really stop exploitation unless the hold bug is fixed first.

Ravendarksky

caffeine... your SS has 5 I pieces incoming.... what bag is that??

myndzi

#25
Quote from: Ravendarksky
caffeine... your SS has 5 I pieces incoming.... what bag is that??

When the sequence is exhausted it just feeds Is. I managed to drop 20+ of them once

caffeine: Ahh, nice. I didn't really consider exploiting the shift delay, most of the rules stayed the same so I somehow thought that would too. Not sure what you mean by the hold bug though - just that it doesn't count for anything?

Here's one that could get you really low: use Paul's 40 lines trick - you can't rotate necessarily, but if you hold shift and soft/firm drop, no inputs are counted and you can stack a bunch of pieces for free. With a bit of intelligent rotating, you can stack them a little cleaner. In the end it usually cost me more KPT than it saved trying to clean up afterwards however. You can exploit the Is too if you stack high instead of low - you run out the sequence and then can use this exploit to stack Is nice and cleanly and get easy clears. Heh! I like your way much better.

Paul676

At first I thought my 40 lines stack would be a viable option, but in fact a better option is just placing pieces free form, and make sure with each piece dropped in an efficient manner that the next piece can also be dropped in an efficient manner. Forcing the move like my stack method does doesn't help the over-all finesse, since it's not for free, the hard drop gives 1kpt.
               Tetris Belts!

myndzi

Quote from: Paul676
the hard drop gives 1kpt.


That's why you hold soft drop and let the lock delay take over

caffeine

#28
Quote from: myndzi
That's why you hold soft drop and let the lock delay take over

One of the nuances I've found with this KPT metric is that it counts any lock as "1 hard drop" key. Whether you hard drop, soft drop, or just let it sit and lock on its own, it'll count it as +1 key. That's probably for the best, anyway. Since, otherwise, I would've just decreased the lock delay and let the piece lock on its own every time.

Quote from: myndzi
Not sure what you mean by the hold bug though - just that it doesn't count for anything?
I believe belzebub mentioned that it was difficult to find how the hold feature was implemented, so he wasn't able to account for it in his KPT metric. As a result, we can hold as much as we want without it adding to our KPT. Another side effect of this is that the keys going towards movement disappear after holding. That's why you'll see in my replay that every time I changed directions, I'd hold that piece I changed directions on and ride it into the next piece.

myndzi

Quote from: caffeine
One of the nuances I've found with this KPT metric is that it counts any lock as "1 hard drop" key. Whether you hard drop, soft drop, or just let it sit and lock on its own, it'll count it as +1 key. That's probably for the best, anyway. Since, otherwise, I would've just decreased the lock delay and let the piece lock on its own every time.
I believe belzebub mentioned that it was difficult to find how the hold feature was implemented, so he wasn't able to account for it in his KPT metric. As a result, we can hold as much as we want without it adding to our KPT. Another side effect of this is that the keys going towards movement disappear after holding. That's why you'll see in my replay that every time I changed directions, I'd hold that piece I changed directions on and ride it into the next piece.


Ah, didn't catch the bit about it nuking your moves up to the hold point, nice.