User talk:Toomai/Moveset tables in progress

Current confusions

1. Despite the fact that I've seen various movesets give Peach's frying pan a sweetspot of 19%, I have never connected with it, and am therefore unable to test it.
2. I am also unable to get a Meteor Smash with Peach's d-tilt.
3. Giga Bowser's down throw is weird. First it does the 20% hit with significant knockback, but it seems like the target and bystanders don't actually take any (they get buried instead). Then, seemingly one frame later (not even at 1/4 speed can I tell the difference between the hits), the 1% hit kicks in and sends the primary victim a fixed distance (so it can't be measured, as the game has already memorized the max launch as the first hit's knockback).
4. The first hit of Yoshi's Egg Throw sometimes misses. (Yes, it does two hits - 1% then 8%.)

Initial discussion

this is very nice. hurry up with bowser's please XD Paper Bowser (talk) 22:20, 18 April 2009 (UTC)

Wow. My only major complaint is the somewhat informal style of the descriptions. Otherwise, though, nice work. (You might want to make the pictures have a description of the type of attack, like your matchup icons did.) Miles (talk) 23:09, 18 April 2009 (UTC)

The pictures will have captions; I just don't want to make them into templates yet. (Using a template with only the image and its caption prevents one of the image's captions from having an error or something.) And yes, the descriptions are informal. I'm not really putting any work into them; they're just placeholders while I do the "real" work. Toomai Glittershine The Stats Guy 02:18, 19 April 2009 (UTC)

Minor question. If F.L.U.D.D. doesn't do any damage, do stickers affect it all? If not, can it really be classified by sticker type? Just wondering. Miles (talk) 22:26, 20 April 2009 (UTC)

Yeah, FLUDD probably isn't affected by stickers (if only because Mario can't wear Water stickers). But it may deal extra knockback to Charizard, and that would cause it to be Water. (However, it may be impossible to get a conclusive test.) Toomai Glittershine eXemplary Logic 23:03, 20 April 2009 (UTC)

Pokemon don't have weaknesses in this game, and almost every attack has multiple types. PK Fire is [Fire] and [PK] AND [Specials:Indirect] There's a whole faq for this on gamefaqs already... - Gargomon251 (talk) 01:48, 21 April 2009 (UTC)

Yes, I know many attacks have multiple types. But I'm guessing that, because Mario can't boost FLUDD with Water, Special:Indirect won't do much either. It will be tested, though.

As for PT's Pokemon...from what I've heard, they take 1.1x knockback when an attack is super effective, and 0.9x when it's not very effective. I'm guessing this only applies to the Fire-Grass-Water triangle, but it's still significant. Toomai Glittershine The Stats Guy 02:31, 21 April 2009 (UTC)

Heard from whom? There's been extensive testing by many people, and I could have sworn that had been debunked- Gargomon251 (talk) 02:59, 21 April 2009 (UTC)

First off, I'm glad to be back. Second, I know for a fact that the PT's pokemon have their weaknesses from the games. There just is no "not very effective" appliance. I've studied Brawl's odd physics for a while now and I've come to that conclusion. Blue Ninjakoopa 05:29, 21 April 2009 (UTC)

I did a minor test just now with an uncharged FLUDD and concluded that it is Water-type. Charizard takes more knockback from it than Mario (even though he weighs a bunch more) and Ivysaur takes less (and he weighs a touch more than Mario). Toomai Glittershine eXemplary Logic 14:26, 21 April 2009 (UTC)

Ye. And moves like Mario's forward smash do a lot more damage to Ivysaur. Blue Ninjakoopa 15:30, 21 April 2009 (UTC)

That is completely false. Mario's Fsmash at point blank does 19% charged to Ivysaur, and it does the exact same to Squirtle. FLUDD pushes Charizard and Ivysaur both exactly the same as well. - Gargomon251 (talk) 17:23, 21 April 2009 (UTC)

I didn't write down the exact values, but an uncharged FLUDD squirt did a low-800s knockback to Charizard, while it did a mid-600s knockback to Ivysaur. The value for Mario under the same conditions is 731. While the values are close enough to be irrelevant in a match, they are still significantly different, and they both follow the 1.1x/0.9x pattern I read elsewhere (I think it was a game physics topic on SmashBoards somewhere).

And BNK, I'm pretty sure that the moves do the same damage. It's just the knockback that gets altered. Toomai Glittershine The Stats Guy 18:01, 21 April 2009 (UTC)

Where are you getting these numbers and formulas? And are you taking traction into consideration as well? And there's two weight listings on this wiki, both of which have debatable accuracy... - Gargomon251 (talk) 22:41, 21 April 2009 (UTC)

That's what I'm trying to say, but Gargomon feels the need to contradict everything I say and gets angry so quickly it's impossible to make reason with him. I'm sorry Gargomon, but I know more about Brawl than you do. >_> Blue Ninjakoopa 18:57, 21 April 2009 (UTC)

That may have been what you were trying to say, but that's not how it came across. You said "a lot more damage", which is not the same as "a lot more knockback". And that was a borderline personal attack there. Toomai Glittershine The Table Designer 20:19, 21 April 2009 (UTC)

The numbers I'm getting are from the Max Launch Speed stat that appears in match results. If you only hit with one attack in the entire match, then the knockback that attack did will be the Max Launch Speed. Also, I'm very sure that traction is not a factor in knockback calculation. As for weight, I only use Mario as the testing sandbag. I do have a weight list I took from SmashBoards, which matches the "Weight 1" list on the Weight page. Toomai Glittershine The Table Designer 23:33, 21 April 2009 (UTC)

i agree with gargomon about the pokemon. if these physics dont apply to pikachu or lucario then they shouldn't apply to three other pokemon. and where are you getting your calculations? Paper Bowser (talk) 01:59, 22 April 2009 (UTC)

What calculations are you talking about here? If you're talking about the 1.1x and 0.9x multipliers for PT's Pokemon, I found them in this topic over on SmashBoards.

As for the other Pokemon not being affected by Pokemon types...I don't know what to say, really. My guess is that, to do that, they would have to set up each move to not only have a Smash type (Arm, Leg, Slash, etc) but also a Pokemon type. As it is, they only look at the Fire-Grass-Water triangle for PT's Pokemon, and to me it's just another way to convince you to use them all equally. Toomai Glittershine The Table Designer 03:01, 22 April 2009 (UTC)

If you want to get some data yourself:

I'm the kind of guy who likes to do everything on his own. However, conducting damage, type, and knockback research on almost 1000 moves is probably not a one-man job. So, here's what to do if you want to contribute data.

Note: If you can hack out the exact data from the game, huzzah! It would save tons of work. I have no idea if it's possible, though. (There's always the hope...)

Damage

Yes, damage values are everywhere. But I check them all myself (in Training Mode to avoid SMN and using 1/4 speed on fast attacks). And some aren't listed, such as floor attacks. I require the damage done by every hitbox of a move. And that means EVERY one - so Marth will have pretty much two complete lists (one tippered, one not). I don't frequent SmashBoards, so I don't know how comprehensive damages are listed there - if there's free data over there, just point it out.

Angle

Basically, the direction the enemy gets sent flying. Easy to test - go to Training Mode, pump it up to 999%, and mark the line of the smoke trail. (1/4 speed highly recommended.) I report the values as such - an attack with an angle of 0 sends the opponent completely horizontal and a 90 is completely upwards. Negative values send people downwards, so test Meteor Smashes and the like on aerial enemies. A "b" (such as 70b) designates that the enemy flies the "wrong way" - instead of flying away from the attacker, the enemy flies past the attacker (and then away, of course). I need an angle for each hitbox.

Type

Does the Mario Tornado do Leg damage, or just Spin (if even that)? It should be simple to test (I haven't done this myself yet, I'm still nailing down the knockback values currently). First, attack an enemy with the move and count how long it take to die. Then slap on a sticker and see if the move does more damage. If it does, the move is of the sticker's type. (Of course, this must be done for each hitbox.)

Knockback

This one is a pain. Use an Angled Special Brawl if you don't want your stats to inflate - this can't be done in Training Mode. Turn Handicaps on and set one guy to the testing character and the rest to Mario (however many you can deal with at once). Set the Mario to the following percentages: 0%, 100%, 200%, and 300%. (Obviously, you can't do all these at once.) Go to Battlefield since there's soft platforms and no hazards.

Now then. Hit the Mario(s) with the testing attack. Then quit the match. Under the match results you'll see "Max Launch Speed". (Ignore "Max Launcher Speed", you don't need it.) Since you only hit each Mario with one attack, this is the knockback that one attack did. For example, if the 200% Mario has a Max Launch Speed of 4910 (ignore the meaningless "mph" units), that means the move did 4910 units of knockback at 200%. Once the testing is done, you'll have four values - the knockbacks the move did at 0%, 100%, 200%, and 300%. Mark down which hitbox the numbers match up with and I'll do the rest.

Sound tough? It gets worse. To test the 100%+ edge attack, you'll have to set the testing character to above 100%. To test floor attacks, you'll need to have the testing character lay down somehow. Either hit him with the dummies or have him carry something big off an edge. You may need to try different things to get both the facedown and faceup attacks tested. For trip attacks, you may want to use Bananas. The good news is, I have reason to believe that both sides (front and back) of a single floor attack are identical. Don't take my word for it, though.

For smash attacks, mark down the values for one that's not charged and one that's fully charged. As for other chargable moves, I've heard that not all of them charge knockback at a predictable rate (such as the Giant Punch). You may have to make a ton of tests for each level of charge (I sure hope not).

Finally, for multi-hit attacks, you have to test the middle hitboxes without connecting with the earlier ones. And shielding, from my experience, alters the numbers somewhat. So you'll have to either dodge precisely or activate the move at just the right height. (The first hitbox is easy enough, just have it connect and then quit the match. Although you have to make sure the hitstun ends first, you may want to use a Timer on the attacker for that.) Luckily, most multi-hit moves have a bunch of identical hitboxes, so you won't have to test every one if they all do the same damage. While I'm not prevented form using data that involves previous hitboxes connecting, you would have to be 100% sure which hitboxes connected and how much damage they do, including any SMN. And no matter what knockback data is collected, there has to be one at 0%.

---

"Use an Angled Special Brawl if you don't want your stats to inflate"....how does camera angle affect stats? What do you mean they "inflate"? Why is this not possible in Training mode? Are you referring to Stale Move Negation? I don't see what else would be impacted in Special Brawl and not in Training... - Gargomon251 (talk) 22:44, 21 April 2009 (UTC)

What I mean here is that, if you use a normal Brawl, all your match exitting and character selections will affect the stats the game keeps such as "Match Reset Count", "Favourite Character", etc. And the reason Training can't be used (for knockback, it can be for damage and angle) is because you don't see end-of-match stats where it says Max Launch Speed. Toomai Glittershine eXemplary Logic 23:33, 21 April 2009 (UTC)

Items

Will they eventually show up? And will you have to cover every possible move with them, as well? Miles (talk) 01:50, 30 April 2009 (UTC)

I'd like to do items, but for now I'm sticking with movesets. (If you do some testing yourself, I'll take the data anyway.) Toomai Glittershine The Table Designer 02:47, 30 April 2009 (UTC)

Request

I know that this is your own page, but could you please link the names in the Universal Language we all know as English? Blue Ninjakoopa 03:15, 5 July 2009 (UTC)

Eh...I'll think about it. Toomai Glittershine The Stats Guy cntrbs 18:58, 10 July 2009 (UTC)

Inactivity Warning

By my current calculations, chances are I won't be able to complete all characters before September. I am going to university at that time, so my ability to do this kind of stuff will be up in the air. Toomai Glittershine eXemplary Logic cntrbs 18:24, 12 July 2009 (UTC)

"Magic" formula....

Hi. I want to post my preliminary moveset table as a user subpage. Could you please tell me what formula you use to calculate the knockback per % damage so I can be consistent with your tables? Oh, and is my new sig fine? I maxed out the number of characters I can use for a sig.... RAN1domchupunch!!! 00:48, October 30, 2009 (UTC)

Sig looks fine. As for the formula...well, it's pretty simple really. Just calculate the slope. Excel does this pretty well. Just put in =slope(rangeOfKnockbackValues,rangeOfPercentages). For example, rangeOfKnockbackValues would be the row that goes something like "2926,3871,4816,5762"; and rangeOfPercentages would be "0,100,200,300" (or whatever you used). You may then have to divide or multiply by 100, depending on wheter your percentage cells are formatted. Toomai Glittershine The Stats Guy 00:57, October 30, 2009 (UTC)

Thanks for the help. I got good values for my knockback per % (I'm using Apple Numbers here; it probably uses the same formula). I'll be posting the table very soon... RAN1domchupunch!!! 01:25, October 30, 2009 (UTC)

On your "Collecting Data" section…

Shouldn't we just use one Mario at a time? If we use multiple Marios, we'll end up having to deal with SMN, and that would be a serious problem. It would lower the strength of the attack, and we'd have incorrect data. We wouldn't want bad data for the project if we want to be accurate. RAN1 14:45, November 28, 2009 (UTC)

It's not much of a problem; just have the attacking character leap off the edge to reset SMN. Using multiple targets means less going in and out of matches, but you have to be careful you don't confuse them or have them run into each other. Toomai Glittershine The Table Designer 14:54, November 28, 2009 (UTC)

So I come across this

and I wonder why it isn't used here. RAN1 01:29, December 13, 2009 (UTC)

Because I haven't gotten around to implementing it yet. No real reason.

Hm, that reminds me, I have the Yoshi data completed but not posted. Better fix that tomorrow. Toomai Glittershine eXemplary Logic 01:34, December 13, 2009 (UTC)

Suppose you wouldn't mind me adding it? RAN1 01:36, December 13, 2009 (UTC)

Eh, go ahead. Toomai Glittershine The Table Designer 01:38, December 13, 2009 (UTC)

Collateral knockback

This is not about a multihit throw (such as Kirby or MK's down throw). I've seen characters be damaged and knockedback by other characters that are reeling having just taken knockback themselves. This effect is noteworthy, as in Sudden Deaths and whatnot this small knockback can cause a KO. Is there any way to test the extent of this effect? (Lava on Norfair was where I saw this the most.) Miles (talk) 00:34, December 20, 2009 (UTC)

Yes, reeling characters can hit each other. It's possible to measure the knockback dealt by the ramming, but since we can't know the speed of the reeling guy at the time of the impact, there's no way to realize a prediction/formula out of it. Toomai Glittershine The Stats Guy 01:16, December 20, 2009 (UTC)

Ah. Ok, thanks for clearing that up. Miles (talk) 02:41, December 20, 2009 (UTC)

Suggestions

This is very useful information, I can see why you are an admin. I do have a couple of suggestions though. You should you use Bowser instead of Mario for your knockback testing. It would be better to use one of the weight extremes instead of the so-called "standard" weight. But this really isn't important and is just my opinion. Secondly, you should find KO percentages as well. The fact of the matter is that a majority of the readers are going to make more use of KO percentages then your knockback rating. For these, using Bowser instead of Mario is even more important as every other character would be KO at or below the percentages found (except for vertical knockback moves but Bowser is still at the top while Mario is in the middle). Bowser is also better for these as his falling speed is in the middle while Mario's is on the floaty side. When my current semester is over, I'll probaly do moveset documentation of my own, but using the suggestions above. I can't do any now however, as I have to finish three papers within the next two weeks. Omega Tyrant 14:35, April 21, 2010 (UTC)

The primary reason for using Mario as the sandbag is because, as far as I know at the present time, his weight is 1.000 and his fallweight is 0.00. This makes calculations a whole lot easier, which is currently more important than providing KO percentages and the like. Toomai Glittershine The Table Designer 16:31, April 21, 2010 (UTC)

I was wondering, exactly how were these weight and fallweight numbers found? Omega Tyrant 17:00, April 21, 2010 (UTC)

This SmashBoards topic contains the knockback formula, as well as the weight and fallweight lists. It's admittedly somewhat old, though, and it's hard to find stuff since searching is currently disabled. Toomai Glittershine eXemplary Logic 17:09, April 21, 2010 (UTC)

Ok then, I can see why use Mario as the sandbag. The link certainly cleared up the questions and doubts I had about the weight chart. However, I still believe finding the KO percentages is important as well and it would be rather simple to find them out. I'll be sure to help you finish these charts when I finish my papers. Omega Tyrant 17:36, April 21, 2010 (UTC)

Hacks

I can hack stuff sort of like this:

.

If you would like I can teach you how to get stuff like this. It is Donkey Kong's dash attack broken down. —Preceding unsigned comment added by highwaypumpy (talk • contribs) 15:48, April 21, 2010 (UTC)

That would be very appreciated - you might want to email me instead of posting it here, depending on the nature of the hacks. It'd also be nice if you could do the same thing for other aspects of the game, such as item attacks, Pikmin frequencies, SSE enemy HPs, and the like (I know nothing about how hard any of that is, though). Toomai Glittershine eXemplary Logic 16:31, April 21, 2010 (UTC)