Printable Version of Topic

Click here to view this topic in its original format

Forums - Hard Drop - Tetris Community _ Strategy/Help _ The 6-3 Split: does it work and is it worth it?

Posted by: caffeine Dec 5 2010, 05:11 AM

Discussion of the 6-3 split has been popping up here and there ever since Maserati used it to beat Jono's 40 line world record*. In theory, it should offer a slight edge, but it's a headache to learn. For that reason, I spent a good six hours gathering and analyzing data today.

Why would the 6-3 offer faster play, anyhow? Well, as Someone2knoe http://harddrop.com/forums/index.php?s=&showtopic=1846&view=findpost&p=14391, it forces you to avoid certain unfavorable placements. Here's why:

In the above diagram, placing pieces in the green columns tend to require the least movement, while the red ones tend to require the most, roughly.

If we position our tetrising column on the 7th file, we can force ourselves to eliminate the possibility of some of these "red column" placements:



By eliminating the possibility of these placements, we inadvertently use less keys.

For my experiment, I transcribed and analyzed over 750 piece placements. I played three games of 100 lines each, using three different methods. I used http://harddrop.com/wiki/0G_60_Hz_SRS_Movement_Finesse, 180 rotation, and didn't use hold. The three methods were: 6-3 split, traditional (making tetrises all the way to the right), and "free form" (not going for tetrises at all). I played 100 lines per game instead of 40 in order to gain more accurate statistics. I played as naturally and logically as possible.

After transcribing the replays to fumen, I recorded how many meaningful keys per tetromino (mKPT) each perfectly-finessed piece required. In the past (when I showed how SRS required less KPT than ARS), people brought up the argument how KPT isn't useful since it doesn't take into account how a player can rotate and move simultaneously. For this reason, I coined (yet another) new metric: mKPT. To calculate it, each movement, rotation, and hard drop counts as one--except when we can execute movement and rotation simultaneously within a single frame. In these cases, mKPT combines them into one key input.

Examples:

1 mKPT (hard drop)


2 mKPT (rotate, hard drop)


2 mKPT (simultaneous rotate+move, hard drop)


3 mKPT (rotate and move events cannot occur simultaneously here)


3 mKPT (move, move, hard drop)


The games

6-3 Split:



mKPT: 2.125

Advantages:

Disadvantages:Traditional:



mKPT: 2.360

Advantages:Disadvantages:Free form:



mKPT: 2.277

Advantages:Disadvantages:Here's https://spreadsheets.google.com/pub?key=0AgsClBPVhGPpdERQUFNmRkVnMENtb2IyTVlVWGl3Umc&hl=en&output=html.

So the experiment backs up the theory, but just how much of an improvement is it anyway? Let's take a hypothetical player who goes at a pace of exactly 10 mKPS. Let's say he tries all three methods out and finishes with a perfect clear in every game. Here's what his records would look like: ((mKPT*100)/mKPS)So, it looks like the 6-3 split definitely has some substance to it. Saving two whole seconds sounds pretty good, right!?

However, I wonder if there's more to it than that. One major criticism I'd like to bring up is how mKPT (and KPT as well) might not be all that important in the first place. Why? First off, the most important factor here is "processing speed." Imagine an interface where the player can "think" each piece into position, 1 by 1, using a total of 0 KPS. Common sense tells us that he probably wouldn't perform at a http://tasvideos.org/ speed all of a sudden. He still needs to think it through, doesn't he? So, what I'm trying to say is that there's likely a "KPT ceiling." Once KPT reaches a certain point, it's likely enough to handle even the fastest player's processing speed. Take for example WR* holders Apocalypse and Maserati. They don't use 2-step finesse or 180 rotation (which would save more keys than a 6-3 split), but they sit at the top of the leaderboard nonetheless. I think think it's a mistake to say that if you drop 0.15 KPT, your 10KPS will stay the same (and as a result gain TPS). In all likelyhood, your KPS will probably drop uniformly until you improve what's actually important--your processing speed. This would explain why we don't see any Keyblox or Typomino WRs yet.

Conclusion? Do what's in your heart! :hug:

*Note: yes, I realize the Koreans have the WR, but we can't see the replay, so what's the point?

Posted by: briannn Dec 5 2010, 05:17 AM

fuk man good job... seriously impressed

Posted by: bach_of_tetris Dec 5 2010, 05:21 AM

I've been trying to do 6-3 with no hold, no skimming, 10 tetrises, and 180 rotate. For me it only seems to work like once every 5 games and my best time is 53 seconds because there is so much looking ahead to do...

If people really want to go for speed and efficiency, then we should start switching to keyblox and practising only that haha

Posted by: Paradox Dec 5 2010, 05:37 AM

Thank you for going through and doing this. Just one thing, can you call it 6-3 ?

3-6 would be this:



QUOTE
This would explain why we don't see any Keyblox or Typomino WRs yet.


There is just a lack of players and the control system is very complicated. It is the most difficult to learn. I hope to prove that it can be much faster than traditional controls if you are able to get used to it.

Posted by: larrytetris Dec 5 2010, 05:44 AM

Did you consider right-side versus left-side traditional stacking? It seems that there are a lot of fast right-side stackers (arf, kdbar), but I'm not sure if there's any actual benefit....

Posted by: caffeine Dec 5 2010, 05:45 AM

QUOTE(Paradox @ Dec 4 2010, 11:37 PM) *

Just one thing, can you call it 6-3 ?

Doh! You're absolutely right. I fixed the in-post stuff, but maybe a mod can change the title?

QUOTE(larrytetris @ Dec 4 2010, 11:44 PM) *

Did you consider right-side versus left-side traditional stacking? It seems that there are a lot of fast right-side stackers (arf, kdbar), but I'm not sure if there's any actual benefit....

Yes, but an analysis didn't seem warranted. It doesn't look like there'd be much difference in theory. I believe we're in the habit of placing the gap on the right because, in high gravity, it offers better mobility. Another thing I considered: placing the gap in one of the two middle columns. However, in theory this would be a bad place to put it (as it eliminates the possibility of hard dropping without needing to move or rotate). Lastly, placing the gap on column 2, 3, or 8, or 9 would also be a bad idea. This is mostly because the islands would be too small to manage (or would eat up all the I-tetrominoes). Not only that, but those are the only places where a vertical I-tetromino requires not 2 mKPT but 3.

Posted by: Paradox Dec 5 2010, 05:50 AM

QUOTE(larrytetris @ Dec 5 2010, 05:44 AM) *

Did you consider right-side versus left-side traditional stacking?


I somehow concluded that right-side stacking was better vs left-side, but I don't remember what my logic was. Now that I am thinking about it I don't know if there would be any difference.

----------------------------
The 6-3 requires some skimming to do it well in my opinion. There are placements that have a good mKPT that you can use to skim over the gap. I don't think its the way to go for TF Sprint.

Also Maserati is basically proof that you can get use to it with time.

Posted by: Noogy Dec 5 2010, 06:01 AM

would've been nice if you had some mKPT counter on your fumens

Posted by: Paradox Dec 5 2010, 06:04 AM

You should go through and do that for him Noogy ;P

Posted by: caffeine Dec 5 2010, 06:12 AM

QUOTE(Noogy @ Dec 5 2010, 12:01 AM) *

would've been nice if you had some mKPT counter on your fumens

In the spreadsheet, the column numbers go along with the fumen numbers. If you find any mistakes, please let me know!

Posted by: Shizi Dec 5 2010, 06:19 AM

most tnet2 players do this

Posted by: crzy242 Dec 5 2010, 06:34 AM

QUOTE(larrytetris @ Dec 5 2010, 05:44 AM) *

Did you consider right-side versus left-side traditional stacking? It seems that there are a lot of fast right-side stackers (arf, kdbar), but I'm not sure if there's any actual benefit....

i dont get a mention? im saddened.

Posted by: XaeL Dec 5 2010, 07:22 AM

all aussies stack on the right. its like how our toilets flush in teh opposite direction.

Posted by: coolmaninsano Dec 5 2010, 07:23 AM

QUOTE(XaeL @ Dec 4 2010, 11:22 PM) *

all aussies stack on the right. its like how our toilets flush in teh opposite direction.


For some reason, I don't think the Coriolis effect applies to Tetris. tongue.gif

Posted by: crzy242 Dec 5 2010, 07:40 AM

QUOTE(XaeL @ Dec 5 2010, 07:22 AM) *

all aussies stack on the right. its like how our toilets flush in teh opposite direction.

me and you are the only two aussie right stackers =[

Posted by: larrytetris Dec 6 2010, 07:41 AM

QUOTE(crzy242 @ Dec 5 2010, 06:34 AM) *

i dont get a mention? im saddened.


Never seen you play before tongue.gif M'bad.

Posted by: chopin Dec 6 2010, 09:24 AM

AWESOMEEEE!!!!!!!!!

Posted by: coolmaninsano Dec 6 2010, 09:33 AM

So is this really worth learning? I can get a left-side stack up at around 2.1 TPS*, will this really improve my speed?

*consistently, can go about 2.4 maximum

Posted by: larrytetris Dec 6 2010, 10:07 AM

QUOTE(coolmaninsano @ Dec 6 2010, 09:33 AM) *

So is this really worth learning?


More importantly, does it matter?

I assume your goal is to get better at multiplayer, having been motivated by suggestions for speeding up in your TTO commentator remarks. If you reduce key presses only in sprint, it's meaningless for multiplayer; you'd have to 6-3 in multiplayer.

In other words, the only improvement in Sprint that makes it worth it in multiplayer would be lower mKPS given the same stacking style.

Posted by: coolmaninsano Dec 6 2010, 10:31 AM

QUOTE(larrytetris @ Dec 6 2010, 02:07 AM) *

I assume your goal is to get better at multiplayer, having been motivated by suggestions for speeding up in your TTO commentator remarks.


Yeah, I am trying to improve my APM in multiplayer. My current strategy (as it seems) is open with a multiple t-spin setup, then just DS through garbage and do random TSDs whenever I can, hoping one will spike my opponent. I do think something like http://harddrop.com/fumen/?m110@XdF3gbI3gbI3gbI3gbI3gbI3gbI3gbI3gbI3gbC3pb?+uH7eLX0Ay3AF6AZdBsMBbLBePBcHB/DBZdBFiBAAA is a better option than what I do now, and I'm willing to take time to learn it, if it's proven best.

Posted by: bach_of_tetris Dec 6 2010, 10:57 AM

QUOTE(larrytetris @ Dec 6 2010, 09:07 PM) *

More importantly, does it matter?

I assume your goal is to get better at multiplayer, having been motivated by suggestions for speeding up in your TTO commentator remarks. If you reduce key presses only in sprint, it's meaningless for multiplayer; you'd have to 6-3 in multiplayer.

In other words, the only improvement in Sprint that makes it worth it in multiplayer would be lower mKPS given the same stacking style.


In multiplayer people only use it in their opening burst, until garbage comes

Posted by: lilasianboy193 Dec 6 2010, 07:20 PM

Thank you for answering my question. I will continue to experiment with what I like. The two tied WR holders use two different styles which was the main confusion for me. I will use this to try to improve my tetris skills more biggrin.gif:D:D

Posted by: Paradox Dec 6 2010, 08:29 PM

@larry not only is this the best for 40L but it is the best tetris stack when working from a clear screen.

I think free-form can be better but it requires some decisive playing which defeats the purpose of free-form.

Posted by: LapSiLap Dec 19 2010, 02:00 PM

Cool topic! Gonna try..

Edit:
yay 42secs biggrin.gif

Edit:
This edit function is so coooool

37..
HoN time.

Posted by: caffeine Dec 19 2010, 04:34 PM

QUOTE(LapSiLap @ Dec 19 2010, 08:00 AM) *

Cool topic! Gonna try..

Edit:
yay 42secs biggrin.gif

Edit:
This edit function is so coooool

37..
HoN time.

New world record in 3, 2, 1...

And no using Tetris Friends! That one's not fast enough.

Posted by: myndzi Dec 19 2010, 04:36 PM

Wow, great work caffeine. I have some things to add.

Free-form has better stats than stacking for tetrises, but you neglected to mention line-clear delay, which exists in many Tetris games. Line clear delay is one of the biggest reasons to go for all Tetrises with no skimming.

I've talked about this before on chats and stuff, but when your goal is to make a 40-lines record, it doesn't really matter how many times you fail. For this reason, even if you can't pull it off once in a hundred times, you can just think of all of those as practice. Keep shunning hold and stacking the best way you can unless it presents a problem for you processing-wise.

And processing is the other reason that what you termed free-form stacking is maybe not as good as its stats claim. I can't quite put my finger on it, but I've always found it to be harder / slower to play that way. I suspect it has something to do with visual perception; dropping pieces and having some of them disappear somehow makes it harder to re-orient, even though the stack shape is no different than if the lines didn't go away. On top of that, there's vertical distance to look between your next pieces. I won't count soft drop speed here since I am talking about the optimal 40-lines game. It may very well be that this, like 6-3 stacking, is just something that can be "gotten over" with practice and work, but I don't know that it would be worth it even if you could do it fluidly... and there's something rather more satisfying about spamming Tetrises anyway wink.gif

Anyway, it would be interesting to see if someone could get good enough at the non-standard "free-form" style play to make significant 40-lines progress, but even your own measurements seem to suggest that 6-3 with just tetrises beats it.

With regards to multiplayer, consistency becomes more important than speed, but on the other hand you aren't stacking 40 lines off the bat. A better argument against using this in multiplayer is that Tetris starts are the least efficient and you'd probably benefit more from learning other openings.

Posted by: LapSiLap Dec 19 2010, 05:19 PM

31.03 now biggrin.gif but bad end.....toook me like 2 secs for the last 2 lines :[

Posted by: Paradox Dec 20 2010, 12:51 AM

Free-form stacking is actually easier for beginners than other methods of stacking (from what I have seen). You probably just aren't accustomed to it after playing with a gap for so long.

I actually think free-form can be more efficient than 6-3. It depends on your luck. or you can literally try practicing using as few key presses as possible.

Posted by: solo2001 Dec 20 2010, 02:28 AM

I will learn this method on keyblox before I die.

Posted by: Paradox Dec 20 2010, 02:33 AM

QUOTE(solo2001 @ Dec 20 2010, 02:28 AM) *

I will learn this method on keyblox before I die.


Be sure to post your records in the 300tPM challenge thread ;x

Posted by: caffeine Dec 20 2010, 02:51 AM

QUOTE(solo2001 @ Dec 19 2010, 08:28 PM) *

I will learn this method on keyblox before I die.

You know none of these methods have any advantage than the other as far as KPT in Keyblox, right? I mean, you were joking right?

Posted by: solo2001 Dec 20 2010, 03:00 AM

LOL! my bad. I kinda zoned out reading this thread icon13.gif

Posted by: LapSiLap Dec 21 2010, 11:55 PM

Uhm 30.98 without this 6-3 build... dunno if this works lol but it kills my brain :/

Posted by: coolmaninsano Dec 22 2010, 12:48 AM

QUOTE(solo2001 @ Dec 19 2010, 06:28 PM) *

I will learn this method on keyblox before I die.


LOL

Posted by: caffeine Jan 13 2012, 04:17 AM

Sorry to necro, but I was reading http://www.time.com/time/health/article/0,8599,2091477,00.html, and it had something interesting to say that I found relevant to this discussion:

QUOTE
The researchers discovered, the more data-dense the average syllable was, the fewer of those syllables had to be spoken per second — and thus the slower the speech. English, with a high information density of .91, was spoken at an average rate of 6.19 syllables per second. Mandarin, which topped the density list at .94, was the spoken slowpoke at 5.18 syllables per second. Spanish, with a low-density .63, ripped along at a syllable-per-second velocity of 7.82. The true speed demon of the group, however, was Japanese, which edged past Spanish at 7.84, thanks to its low density of .49. Despite those differences, at the end of, say, a minute of speech, all of the languages would have conveyed more or less identical amounts of information.
"A tradeoff is operating between a syllable-based average information density and the rate of transmission of syllables," the researchers wrote. "A dense language will make use of fewer speech chunks than a sparser language for a given amount of semantic information." In other words, your ears aren't deceiving you: Spaniards really do sprint and Chinese really do stroll, but they will tell you the same story in the same span of time.

I believe the same thing happens with Tetris: there's a tradeoff between keys per tetromino and keys per second (in that it is analogous to the tradeoff between syllable-based average information density and the rate of transmission of syllables). So, while you're saving KPT by learning full 2-step finesse, you'll probably compensate by slowing down in KPS and play at the same TPM (at least until you learn to actually play faster). However, that's not to say there aren't advantages in learning 2-step.

Posted by: Paradox Jan 13 2012, 04:23 AM

Well sometimes when you are stacking there are 2 piece or 3 piece patterns that you see and can pretty much stack as fast as you can, having good finesse speeds those patterns up. I figured that out when i played keyblox and sometimes i would place 2-3 pieces super super fast and then continue with my regular pace afterwards.

I'm a pretty strong believer that lower KPT is very beneficial.


Posted by: myndzi Jan 13 2012, 09:28 AM

Seems to me that what he's saying is more along the lines of "your mental speed is more of a limiting factor than your physical speed" - which I can get on board with. Of course it's beneficial to have optimal finesse, but his point is that on the average it may very well not matter if you stack three pieces faster than someone else can stack the same three pieces - since you are still going to be processing the same information at the same rate (assuming equal skill).

Another demonstrative example of this is how there are many things that we can agree are good ideas in general, like rotating both directions, using finesse, balancing your key bindings, etc. And yet, some of the fastest players break one or more of these "rules of thumb". I really do think that mental speed is the primary limiting factor; I can't speak for faster players, but I have certainly never been limited by my ability to input key commands (i.e. fingers aren't moving as fast as I need them to) - and even if this is the case, physical speed is the easiest to improve through practice. Witness some of the freaking insane tap rates of speed runners; while I was watching the SDA marathon recently, I think somebody commented that one of the players could mash 16 times per second. That's crazy, and it's also far faster than anybody can "think" Tetris currently.

Posted by: Paul676 Jan 13 2012, 01:32 PM

myndzi's right that it's about physical speed vs mental speed. For speaking, since people are adept at speaking since they've done it often from an early age, it would be mental speed which would be the limiting factor.

For tetris players, since most of us have played for less than 5 years even, and certainly games which have allowed us to go at sub-30 speeds have been around for less than 5 years, it is generally our mental speed slowing us down. But for awkward piece placements where with bad finesse you'd have to tap 3 times and rotate, or something of that order, it might be better to use some finesse...but something like das to wall and rotate comparated to das to wall+rotate at same time, then tap back has 1 more kpt but no more time. This is, I'd argue, a reason why people who seemingly break the rules do better; 2 taps (das then tapback) and 2 rotates at the same time=4kpt, but only 'uses up' 2 key-times. I'm tempted to re-write the rules of 2 step finesse based on this, but would far rather that someone else did. It seems more correct than normal finesse in terms of time spent...and quicker in terms of time learning it because it's more in line with what we naturally do anyway.

Posted by: caffeine Jan 13 2012, 03:28 PM

QUOTE(Paul676 @ Jan 13 2012, 07:32 AM) *
but something like das to wall and rotate comparated to das to wall+rotate at same time, then tap back has 1 more kpt but no more time. This is, I'd argue, a reason why people who seemingly break the rules do better; 2 taps (das then tapback) and 2 rotates at the same time=4kpt, but only 'uses up' 2 key-times. I'm tempted to re-write the rules of 2 step finesse based on this, but would far rather that someone else did. It seems more correct than normal finesse in terms of time spent...and quicker in terms of time learning it because it's more in line with what we naturally do anyway.

Let's get something clear. You're referring to this move:

And you're saying we should change it in the wiki to this move:


While I'll agree that it's the move more people sort of "grow into" over time, I think that's only because the first one is less obvious. I think there are something good reasons why we should teach method #1 on the wiki, and not #2. First off, it's less work, more precise, and better suited for the wiki. Secondly, in these maneuvers, which we use on the left and right wall with all pieces but O, method #1 is definitely preferable to #2 when you encounter cases where piece A is ARed to the wall, then piece B is ARed to the wall immediately afterwards (where left or right his held down the whole time). In games where this results in piece B instantly appearing at the wall after we drop piece A, it causes extra time for method #2 since they're used to rotating "towards" the wall and then moving back. In method #1, there is no moving back, only the rotation (away from the wall). It's better to illustrate it:




Edit: I'm looking over this again, and one thing you might say is that technically you can rotate and move back at the same time with method #2. While this is true, I still don't think it's a good reason enough to change the wiki. Just because it can be done, doesn't mean it's always like that in practice. Besides, I see the wiki's page as a outline for optimal finesse. You could make a separate page for a more streamlined learning system, but to me those sort of moves are the same ones that people learn by themselves in the first place.

Posted by: Ravendarksky Jan 13 2012, 04:41 PM

caffeine while I am almost in agreement with you I DO have some questions...What about if you are DASing another piece afterwards..When I try and do what you are suggesting, sometimes I end up doing the following:



I would argue that tapping back, while slower. Can be more precise in these sort of situations. Am I just doing it wrong?

Posted by: Paradox Jan 13 2012, 04:56 PM

QUOTE
I would argue that tapping back, while slower. Can be more precise in these sort of situations. Am I just doing it wrong?


I wouldn't say more precise at all. It seems like you just aren't used to it if you make that particular mistake. If i did something like that then I would continue practicing the movement until i got it right.

Posted by: caffeine Jan 13 2012, 05:08 PM

QUOTE(Ravendarksky @ Jan 13 2012, 10:41 AM) *

I would argue that tapping back, while slower. Can be more precise in these sort of situations. Am I just doing it wrong?

I used to have that problem sometimes at first, too. You're holding down the "move right" button through the rotation. You need to let go of that button more quickly. It's just a matter of getting used to the timing. This never happens to me anymore.

Posted by: Paul676 Jan 13 2012, 05:29 PM

*lalala rotate+tap-->tap* causes me no difficulty and is 2 moves Grin.png

p.s. "I'm looking over this again, and one thing you might say is that technically you can rotate and move back at the same time with method #2" This.

I'm not talking about changing the wiki...just adding a new section to it (whenever that may be, likely never)

Posted by: Ravendarksky Jan 13 2012, 07:43 PM

QUOTE(caffeine @ Jan 13 2012, 03:28 PM) *

method #1 is definitely preferable to #2 when you encounter cases where piece A is ARed to the wall, then piece B is ARed to the wall immediately afterwards (where left or right his held down the whole time).

I just don't see how this is EVER possible seeing as you are telling me I'm letting go of the key too late..... Surely if you hold right the whole time both pieces will always end up at the wall

Posted by: caffeine Jan 13 2012, 07:47 PM

QUOTE(Ravendarksky @ Jan 13 2012, 01:43 PM) *

I just don't see how this is EVER possible seeing as you are telling me I'm letting go of the key too late..... Surely if you hold right the whole time both pieces will always end up at the wall

Huh? In this case, it's the second piece that you'd need to let go of the key when rotating, not the first.

Posted by: Ravendarksky Jan 13 2012, 08:00 PM

QUOTE(caffeine @ Jan 13 2012, 07:47 PM) *

Huh? In this case, it's the second piece that you'd need to let go of the key when rotating, not the first.



So you can't use this method to put a piece in the second last column while holding right to allow the next piece to go straight to the wall too?

Because if this isn't the case then I don't understand the advantage over tapping back.

Plus if you TAP back you gain the advantage on the other side of the stack:




Posted by: caffeine Jan 13 2012, 08:27 PM

QUOTE(Ravendarksky @ Jan 13 2012, 02:00 PM) *

So you can't use this method to put a piece in the second last column while holding right to allow the next piece to go straight to the wall too?

Because if this isn't the case then I don't understand the advantage over tapping back.

I'm not really talking about what happens with the first piece. All the matters is that it goes to the wall and preserves the DAS so that the next piece instantly appears at the wall. After that happens, with method #2, you need rotate and move off the wall, but with #1 you just need to rotate.

QUOTE(Ravendarksky @ Jan 13 2012, 02:00 PM) *

Plus if you TAP back you gain the advantage on the other side of the stack:



The "instantly appear at the wall" trick only works once the DAS charge has been preserved from the last piece. Since there's no DAS with that move, there's no preservation and it won't result in having the next piece appear instantly at the wall.

Posted by: XaeL Jan 13 2012, 09:01 PM

QUOTE(Ravendarksky @ Jan 13 2012, 04:41 PM) *

I would argue that tapping back, while slower. Can be more precise

This is never the correct argument for putting something in the wiki. You want to argue that it is FASTER than 2 step finesse in some way.

I however do have some arguments.

Here is a diagram with 5 frame DAS charge and IRS. Using 2step vs das-tapback.

What i about to demonstrate will blow your feeble minds.



DAS tapback.



Now some additional points:

DAS tapback is more lenient and flexible -> HOLD RIGHT, tap LEFT. If u want to keep charge for next piece, keep holding right. If you dont, release right. If you want to charge LEFT, hold LEFT.

Posted by: caffeine Jan 13 2012, 10:15 PM

QUOTE(XaeL @ Jan 13 2012, 03:01 PM) *

DAS tapback is more lenient and flexible -> HOLD RIGHT, tap LEFT. If u want to keep charge for next piece, keep holding right. If you dont, release right. If you want to charge LEFT, hold LEFT.

What game is this supposed to work for? In NP, it allows you to tap back while holding down RIGHT, but it resets the DAS charge. Do I need to tweak a setting or something?

Posted by: XaeL Jan 14 2012, 01:51 AM

QUOTE(caffeine @ Jan 13 2012, 10:15 PM) *

What game is this supposed to work for? In NP, it allows you to tap back while holding down RIGHT, but it resets the DAS charge. Do I need to tweak a setting or something?

My bad it does reset das charge.

Posted by: Paul676 Jan 14 2012, 05:48 PM

does hold reset das charge if you have right held down? That way you can make it faster by using hold :O

Posted by: caffeine Jan 14 2012, 05:53 PM

QUOTE(Paul676 @ Jan 14 2012, 11:48 AM) *

does hold reset das charge if you have right held down?

Nope.

Posted by: Paul676 Jan 14 2012, 05:58 PM

Does something like this check out? I must be close...


Maybe 1 more frame for the I...anyway here is a clear case where hold is better for speed Grin.png

Yes I'm bad at fumen...get over it

Posted by: caffeine Jan 14 2012, 08:28 PM

QUOTE(Paul676 @ Jan 14 2012, 11:58 AM) *

Does something like this check out? I must be close...


Maybe 1 more frame for the I...anyway here is a clear case where hold is better for speed Grin.png

Yes I'm bad at fumen...get over it

Yep, in that particular case, using hold saves frames. I'd argue that the DAS frames saved in the rare occasions where we do this are not worth the frames wasted overall in "hold-oriented play." To me, the advantages of not using hold are obvious, and it just seems like you're trying to justify the style you've already grown accustomed to. =b

Posted by: Paul676 Jan 14 2012, 09:25 PM

to me, you're just trying to defend the style you have taken so long to learn, so that you don't have to feel all that effort is wasted =b

Posted by: caffeine Jan 14 2012, 09:30 PM

QUOTE(Paul676 @ Jan 14 2012, 03:25 PM) *

to me, you're just trying to defend the style you have taken so long to learn, so that you don't have to feel all that effort is wasted =b

Tell ya what. Show me a 40L game where you effectively use this technique to save more frames than you waste by using hold, and I'll happily give it an honest effort to try to learn and use it myself. ;]

Posted by: Paul676 Jan 14 2012, 09:46 PM

I believe the 2 holds at the first 20 seconds of http://www.youtube.com/watch?v=PYk0k7vsVXY (and certainly the first hold of the I) save me keypresses. Then I hold the O needlessly and thereby screw up later on.

But notwithstanding that, surely it's more about the definite possibility of this working (as shown in the last fumen) that should be persuading you to have ago, not whether Paul676 who is crap at finesse can do it once in a blue moon?

Posted by: caffeine Jan 14 2012, 10:04 PM

Do you have a .rep file?

QUOTE(Paul676 @ Jan 14 2012, 03:46 PM) *

But notwithstanding that, surely it's more about the definite possibility of this working (as shown in the last fumen) that should be persuading you to have ago, not whether Paul676 who is crap at finesse can do it once in a blue moon?

Yes. Only, as I was talking about before, I think using hold probably wastes more frames than it saves. Suppose I play a game and use 300 key inputs and finish in 30 seconds. That means, on average, every key stroke costs me 0.10 seconds. I may've counted wrong, but in your video you use hold roughly 10 times. Suppose I play another game with hold, skip DAS on two pieces by using hold, and use hold a total of 10 times. I'll save ((5 DAS)*(1/60 frames))*2 = 0.1666 seconds by doing the trick. I'll waste (0.10 seconds per key)*(10 holds) = 1 second for the 10 hold inputs. That nets me a total of 0.1666 - 1 = -0.8333 saved seconds. In other words, in theory it wasn't worth it.

Posted by: Paul676 Jan 14 2012, 10:53 PM

In the video I use hold 6 times...I messed up at the end because I realised I was onto a record time - but what I mean by the technique isn't relying on hold (as I currently do), but it is recognising whether it would be beneficial to use hold and then using it if it is beneficial.

Posted by: Sisu Apr 21 2018, 12:25 AM

yes

Powered by Invision Power Board (http://www.invisionboard.com)
© Invision Power Services (http://www.invisionpower.com)