Input Delay in Brawl

[DeLux]

Could be a large reason why we don't have a single ICs player in the world that literally never drops grabs

There are some better at it than others, but generally claims like "X player NEVER drops grabs" is hyperbole.

Even in the best of practices for hours and days on end I'll only hit like 96% completion on stocks with tripless CGs

 

[1PokeMastr]

I don't see how this would be a massive game changer. Or make anything terrible happen at all to the game.

As much as Reflex is arguing about it, it's not a big deal, at all.
Both players would receive the 0-2 frame input lag, not just one. This doesn't make the game luck based at all. Honestly, No one knew about it in the past, no one complained about it. There hasn't been ANY dire consequences from this at all.

Given the fact that most, if not ALL attacks have a lingering hitbox. I'm positive there's no move in the game where the hitbox lasts for just one frame. Making the "missed punish" argument.. kind of worthless.

There's a 0-2 frame delay that's been around since 2008 for Brawl and it's just being noticed now. For all of you who are arguing about this, look at yourselves, You've been playing the game for almost 5 years (Assuming you got it when it came out).


I don't see how this is a game changer if it affects both players, hitboxes last for more than one frame.
Quit "crying" about this.

Hell, because of this. Mk's Uair might even be frame 0, frame 1, or frame 2.
Complain about how broken this Make's Metaknight, please.


Also the buffer window invalidates this on most things.

 

[DeLux]

Nope, ICs are bad is all this would mean

And the idea of "power shield everything" is not a viable strategy

 

[TheReflexWonder]

No, you're wrong about it being negligible. I would recommend reading my previous posts to show how prevalent this issue is in normal gameplay.

 

[1PokeMastr]

I read all your posts, and I got tired of reading them because they didn't serve a point. Except for the fact a four year old "coding error", being made known to everyone in 2012, after it's been out for this long.

Is suddenly the turn of the century for playing brawl.

 

[TheReflexWonder]

It's not an issue for most people, as many people have embraced the slow-going spacing there currently is.

Now even people will opt to go with a proactive approach simply because most of them are literally forced to be unreliable. That takes all the wind out of my sails in trying to find ways around the current metagame.

 

[1PokeMastr]

So something we can't see by eye, and have been used to for 4 years, makes this a huge deal ?

 

[TheReflexWonder]

It's something I see every time I get my well-timed attack powershielded by a Dedede or Falco fastfall airdodging to the ground. Instead of getting the reasonable punish I deserve, I take 40% and possible death. I have a major problem with that. I imagine some people don't find themselves in that position often, but I see it an awful lot.

The reason I was able to live with it is because I was under the impression that I just needed to time things better. Now a whole plethora of what-should-be-reliable options are no longer available to me, and any attempt to make use of them in those situations is calling upon Lady Luck to make or break it.

Powershielding will always look like the unfortunate side-effect of this now.

 

[infiniteV115]

I guess I'm lucky to play a character whose landing punishes are a huge grab that's out for 14 frames and a gigantic transcendent hitbox that's out for 5 frames and is safe on PS

 

[1PokeMastr]

Mario's Up B Oos says Hello :3

 

[infiniteV115]

He's still Mario tho.

 

[Elyssa Xey Hexen]

lol The biggest thing I got out of this thread.

Magus will have to work harder at hacking brawl for us.

 

[Biz_R_0]

With all due respect (jk) no one cares about your beliefs.

With all due respect (none) you're being a ****.

You just have to follow the SF scene to see and aknowledge that people can consistently perform links that are 1 or 2 frames tight.

Consistently as in most of the time or consistently as in all the time? Because I did follow SF4 for a while and there were many drops to be had.

And I'm afraid Reflex is right, a very important part of Brawl is punishing landings and spotdodges and when the input delay has a chance to screw 50 to 100% of your window the only thing I can say is : Sakurai.

So what's left to do, abuse lingering and mutliple hitboxes ? Gatling jab sequences ?

Or just do something that lasts longer than two frames, there are a lot of them.

Except for the fact a four year old "coding error", being made known to everyone in 2012, after it's been out for this long. Is suddenly the turn of the century for playing brawl.

This. This guy.

 

[bubbaking]

I pop back in here and I see a bunch of players preparing their johns. If you play this game, you don't have an excuse. I've lost matches to trips and I johned about it non-stop, but in the end, I had to admit that it was my fault for playing this game, and I still play it, as dumb as it is.

 

[Death Arcana]

I agree with reflex 100%
U can't really hard punish the best spotdodges consistently
Don't tell me multi hit moves that defeats the purpose of a "hard punish"
It's either camp for days
Or take a chance and go in just to take another chance on whether or not your attack will be delayed

 

[Vkrm]

The fact that people enjoy the game is grounds for continuing to play it. You already forgiven so much that this is just one more thing. I feel bad that you guys CAN forgive this, but the less I say the better. I'm pretty sure I can say melee>brawl in a competitive sense and now no one can really disagree.

 

[Biz_R_0]

Vkrm you don't even play Brawl, why do you even care?

@DKS not true, there are still good moves that are active for longer than two frames, aka almost all of them.

 

[bubbaking]

I never disagreed and it's true. Competitive brawlers who call themselves competitive brawlers (including myself) have already committed to acknowledging, 'forgiving', denying, ignoring, accepting, and/or fighting through a lot of crap in Brawl, because in the end, they want the money, the fame, the competition, or the fun that the game brings. I find Brawl fun, even though I personally think it's a ton of garbage. I also try to play games like DBZ Budokai competitively, so that says a lot about myself.

Edit: Buffer isn't always good; it can cause some stupid things to happen (e.g. You're doing a grounded move and somehow, in the early frames of that attack, you end up in the air and that grounded move gets turned into an aerial that you never wanted), but we learned to use it and work around it. Tripping is obviously anti-competitive, but if you play this game, you've accepted it. Hitstun cancelling removes combos, so we fish for whatever combos we can pull off while adapting the rest of our play to capitalizing off of isolated single hits. Improved defensive mechanics promote extremely defensive/campy play, so we create rules to deal with ledge planking, scrooging, and all other kinds of ridiculousness. Yes, this game has now added another bad feature to the list. We now know about it, but we can't do anything about it. I hate to be THAT guy, but it's not really productive or worthwhile to be arguing about it like this. It this feature is the last straw that makes one want to drop Brawl, then drop it. If you still want to keep playing this bad game, then prepare yourself, 'cause as Taskmaster would say, "You're in for a world of hurt!!!"

 

[infiniteV115]

I agree with reflex 100%
U can't really hard punish the best spotdodges consistently
Don't tell me multi hit moves that defeats the purpose of a "hard punish"
It's either camp for days
Or take a chance and go in just to take another chance on whether or not your attack will be delayed

It doesn't have to be a multi-hit move, just one that lasts a decent amount of time.
eg Falco's spotdodge lasts for 22 frames and is invincible on frames 2-20.
Take ZSS' dtilt for example. Its hitbox is out on frames 5-7.
This means that assuming there are 0 frames of delay, she can start her dtilt anywhere between frames 15 - 18 of Falco's spotdodge and she'll hit him. That's a 4 frame window. To START on frames 15-18 with a 0-frame delay, this means that she'll have to input it on frames 14-17 of Falco' spotdodge (with 0 delay, the move starts 1 frame after the frame on which it was inputted).

Now yes, assuming that there's a 1 frame delay, then inputting on frame 17 would lead to dtilt STARTING on frame 19, which means the hitbox comes out on frame 23 of Falco's spotdodge (ie his shield is up now). So you get ****ed over on that side, but it also means you can input on frame 13 (1 frame earlier than usual) so you can also fortunately hit them even if you mistime your input.
So with 1 frame delay, your input window would be frames 13-16 rather than 14-17.
With 2 frame delay, your input window would be frames 12-15.

Notice how inputting on frames 14 or 15 of Falco's spotdodge will guarantee that you hit Falco with ZSS' dtilt no matter what the frame delay is, and inputting on frames 12/13/16/17 will occasionally cause you to hit him (depending on the frame delay). And this is with a move that only lasts for 3 frames.

...Is it really that bad to have a 2 (or more) frame punish window on a move with 2 frames of vulnerability? I really don't think it's that big a deal. Most characters have a quick move that lasts for LONGER than 3 frames (ie hitboxes are out for longer than ZSS' dtilt), meaning they have an even bigger, guaranteed punish window.


Dodges are supposed to be hard to punish anyway when used properly -_-

 

[Rikana]

Fun fact: input lag affects everything. Including the menus.

 

[MegaRobMan]

Trying to grab Dedede out of a spotdodge sounds like a bad idea for just about every character except Olimar btw

Except for people who are good enough to grab people out of spotdodge and would be without this crap. It's honestly not that difficult IN MELEE to grab someone after they spot dodge. Hell you can even JCgrab someone if you feel like it.

This is a big deal, and it means I have lost an argument with my fellow melee player brethren, who have mentioned to me before that they felt that there inputs were lagged and I told them they sucked and to go complain more about random tripping.

Melee is a game where you use every frame and can count errors based on how many frames you were off, so people who have been playing that for 8+ years competitively, and switch to brawl, are going to hate this slow, unresponsive game.

I HAVE LOST MATCHES TO THIS, imo.

This is far worse than tripping as far as how much it affects the flow of the game, I think (mostly because of prevalence, but, it is what it is).

I don't agree with this, lol.

RANDOM tripping without being hit by a move is probably the worst idea ever. Tripping due to bananas or zeldas d-tilt I'm still okay with.

no **** tripping.

I hope he meant get rid of RANDOM tripping, I don't support getting rid of tripping, that would make diddy mid-tier.

 

[Flayl]

Spot dodges in Melee are a lot worse...

 

[Deleted member 189823]

Where can I find the hack for normal Brawl? (not Project M)
I'd like to try it myself to see if it makes much of a difference. I am commonly known for having troubles with general game lag the most.

 

[Rikana]

The code has not been released to public.

 

[dettadeus]

Alright, I just did a bunch of testing in Dolphin's frame advance with Ice Climbers (Jab) and Meta Knight (Dtilt).
I'm assuming there's just a flat delay for inputs in Dolphin, since the soonest I could ever get an input to come out was the 2nd frame after the frame I input the attack on.
The way I counted was this (I made the attack input for only the first frame, then released it the second):
Frame 1: Attack Input
Frame 2: No Attack Input 1
Frame 3: No Attack Input 2 (earliest attacks could come out for me in Dolphin) [Counting as 0 Frames Lag]
Frame 4: No Attack Input 3 [One Frame Lag]
Frame 5: No Attack Input 4 [Two Frame Lag]
Frame 6: No Attack Input 5 [Three Frame Lag]

I got by far the largest number of attacks coming out with one frame of lag, nearly double the number I got with any other frame, with both Ice Climbers and Meta Knight. At about half the number was 0 frames of lag, and I only got one or two attacks coming out after two frames of lag.
Oddly enough, one time MK's dtilt came out after No Attack Input 5, meaning it would've had 3 frames of lag, but I'm discounting that as an error for now.

What I found most interesting was that it wasn't particularly random. There would be long sequences of attacks where there were 0 frames of lag, then a large sequence with 1 frame, then another long sequence with 0 frames, etc. There were a few instances where there were only two or three consecutive attacks with the same amount of lag.

Also, the few times I actually looked for it, Nana always had the same number of lag frames as Popo did; her attacks would always come with her usual input lag after Popo's lag was taken into account, although if I tested more it's possible there could be some instances where Nana lags after Popo does.

EDIT: Just tested a bit more. Still getting 1 frame a majority of the time regardless of the animation (I tested MK's shield this time) but I also got... A chain of 8 inputs stuck at 2 frames of lag doing Icies shield, followed by a 0-frames lag shield.

Hope to god you don't ever randomly get 2 frames on one of your inputs because it's going to stick for a while.

 

[shanus]

Alright, I just did a bunch of testing in Dolphin's frame advance with Ice Climbers (Jab) and Meta Knight (Dtilt).
I'm assuming there's just a flat delay for inputs in Dolphin, since the soonest I could ever get an input to come out was the 2nd frame after the frame I input the attack on.
The way I counted was this (I made the attack input for only the first frame, then released it the second):
Frame 1: Attack Input
Frame 2: No Attack Input 1
Frame 3: No Attack Input 2 (earliest attacks could come out for me in Dolphin) [Counting as 0 Frames Lag]
Frame 4: No Attack Input 3 [One Frame Lag]
Frame 5: No Attack Input 4 [Two Frame Lag]
Frame 6: No Attack Input 5 [Three Frame Lag]

I got by far the largest number of attacks coming out with one frame of lag, nearly double the number I got with any other frame, with both Ice Climbers and Meta Knight. At about half the number was 0 frames of lag, and I only got one or two attacks coming out after two frames of lag.
Oddly enough, one time MK's dtilt came out after No Attack Input 5, meaning it would've had 3 frames of lag, but I'm discounting that as an error for now.

What I found most interesting was that it wasn't particularly random. There would be long sequences of attacks where there were 0 frames of lag, then a large sequence with 1 frame, then another long sequence with 0 frames, etc. There were a few instances where there were only two or three consecutive attacks with the same amount of lag.

Also, the few times I actually looked for it, Nana always had the same number of lag frames as Popo did; her attacks would always come with her usual input lag after Popo's lag was taken into account, although if I tested more it's possible there could be some instances where Nana lags after Popo does.

EDIT: Just tested a bit more. Still getting 1 frame a majority of the time regardless of the animation (I tested MK's shield this time) but I also got... A chain of 8 inputs stuck at 2 frames of lag doing Icies shield, followed by a 0-frames lag shield.

Hope to god you don't ever randomly get 2 frames on one of your inputs because it's going to stick for a while.

Glad you confirmed our findings. if you do it with other types of controllers (i.e. wiimotes) it can go even higher

 

[Biz_R_0]

Fun fact: input lag affects everything. Including the menus.

This is why I think this was a bug, because there would be no reason to include this in the menu if this really was a case of "Sakurai hates us".

 

[Ishiey]

Good info, dettadeus. Updated the OP with your test results.

 

[Player-1]

I hope he meant get rid of RANDOM tripping, I don't support getting rid of tripping, that would make diddy mid-tier.

nope, bananas are slips not trips noob

 

[Kink-Link5]

I like how this thread managed to go to **** by post 4. Good stuff.

Also surprised at how long it took to mention Ice Climbers CGs.

Reaction time randomly fluctuating between 12 and 14 frames is no joke.

 

[Biz_R_0]

Hopefully people will stop using ICs so much now that we know CG isn't very reliable anymore.

 

[bubbaking]

There is hope! ICs may not always win from three mistakes!

 

[Player-1]

hey




diddy has infinites too you know

 

[dettadeus]

Alright, after doing a bunch more testing, it would appear as though the lag isn't particularly random at all.

I've found it impossible so far to get two characters to have different amounts of lag when making an input on the exact same frame. When testing with two Falco's jabs, they always started at the same time and clanked, regardless of how many frames of lag happened to be in effect (to phrase it more understandably, it didn't matter if there were 0, 1, or 2 frames of lag, they always experienced the same number when inputting on the same frame). At first I thought the number of frames of lag loosely depended on how many inputs the player had made (as if there was a queue) so I then had one Falco jab and shield a couple times without the other one. After that I went back to clanking, and no matter what they still experienced the same amount of lag when inputting on the same frame. I also tested with one Falco jabbing and the other shielding, and they still experienced the same lag, even when I delayed the Shield input by one frame to catch the 1st powershield frame.

I then thought this might just be character specific; that each character is affected by a different amount of lag when inputting on the same frame, but if two of the same character makes an input on the same frame, they experience the same lag.

Switch P2 to Sheik, P1 is still Falco.

Exact same results with Jab and Shield, regardless of which was doing which.

Then I tried having Sheik Dsmash (hitbox out Frame 4) and Falco Powershield with the first shielding frame. They still experienced the same number of lag frames, despite Falco making his input on Sheik's third frame of input, and Falco did indeed powershield the Dsmash on the first frame. I repeated this successfully two more times.

I also had Sheik and Falco jab with 1 frame of difference in their inputs, and consistently had them start their jab 1 frame after each other, implying equal input lag.

I'm absolutely certain the lag has to be different at some point because there is some frame on which the amount of lag changes, but I haven't actually found evidence of this yet. I'm gonna go keep testing.

 

[Biz_R_0]

The plot thickens!

Seriously, though, I really hope both players experience the same lag, as that would make spot-dodge roulette a non-issue.

 

[#HBC | Joker]

so what you're saying is that in a vast majority of situations, one player will not be screwed over by getting 2 frames of lag when reacting to someone else's input done with no lag. It's not entirely random, even if it is slightly variable from situation to situation, it'll always affect both players equally.

That's good to know.

 

[dettadeus]

I'm still doing a bunch of testing, I've only gotten one instance out of 9 tests of Falco 1st frame PS vs Sheik Dsmash where Falco didn't PS on frame one (he did on frame 2), which is either me accidentally delaying the shield input or them finally experiencing different lag.

Right afterwards, I had them both jab and they clanked with the same lag though, so it's probably more likely I screwed up.

I'm going to start messing around with laggier attacks (sheik's Usmash is next on my list) to see if I can bring up larger frame differences.

After that I'm going to screw around with save states to see if I can get lag to be different on an exact frame by repeating that frame over and over.

 

[bubbaking]

it'll always affect both players equally.

That's good to know.

Like your chances at tripping.

 

[dettadeus]

Like your chances at tripping.

Except this isn't random the same way tripping is, as far as I can tell.
This would be like if, when you trip, your opponent trips as well no matter what they're doing, even while in the air.


Also I've been completely unable to get characters to undergo different amounts of lag, even with different attacks vs shield. Consistently PS'd Sheik's Usmash (hits frame 11) with the same frame inputs 5 times in a row, as well as Falco's Dsmash (hits frame 7) 11 times in a row... each time with the exact same amount of delay for both characters.

I'm going to go ahead and bet before I do this that I'll be able to hit D3 out of his spot-dodge 15/15 times with MK's Dtilt.

 

[Zankoku]

Does anyone know if this negatively impacts DACUS? Especially the tighter ones like for Sheik. It would certainly help explain why there's actually nobody in the world who can do Sheik's DACUS at 100% success rate.