• Welcome to Smashboards, the world's largest Super Smash Brothers community! Over 250,000 Smash Bros. fans from around the world have come to discuss these great games in over 19 million posts!

    You are currently viewing our boards as a visitor. Click here to sign up right now and start on your path in the Smash community!

SSB64 Data Collection

Superstar

Smash Champion
Joined
Feb 9, 2007
Messages
2,351
Location
Miami, Florida
I got bored, and I have a looooot of free time on my hands. I need to finish some work and study for the college finals, so since I won't be able to finish, I'll publish my current findings. I don't have antD's tools, but I tried to find the best I could with what I had. Which means a controller and Mupen video recorder, along with CamStudio's Frame by Frame Player.

Move Degredation
In SSB64, moves degredate in one hit. So after using a 14% move, for example, next use will be 11%, and the next will be 11% too. After some tests, the formula is:

Formula: x is original attack
Deg = [x/4] [] means greatest integer. In other words, round down
New attack = x – [x/4]

Number on left is the original attack, number on right is decrease, number in [] are regains. For most attacks in the game, you need 3 hits to regain the move. Nothing degrades below 3, if you plug it into the formula you get the original attack. I haven't calculated regains yet.

Degrades
4: 1
5: 1
7: 1 [0 1]
8: 2 [1 2]
9: 2
10: 2 [1 1 2]
11: 2 [1 1 2]
12: 3 [1 2 3]
13: 3 [1 2 3]
14: 3 [1 2 3]
15: 3 [1 2 3]
16: 4 [2 3 4]
17: 4 [1 3 4]
18: 4 [1 3 4]
19: 4 [1 2 4]
20: 5 [2 3 5]
25: 6 [1 3 5 6] ?
30: 7 [2 4 6 7]

These were gotten by jotting down attacks. I'm not releasing my current damage list yet, wait till its done. XD

Off the top kills [Hyrule]:
Off top death tests [damage from before Mario USmash taken]. Since I’m unsure how knockback is calculated, I put damage from before and after the USmash. Tested on castle base floor [as in the main part where you start]:

DK: 97 116
Falcon: 87 106
Link: 86 105
Yoshi: 85 104
Fox: 80 96
Samus: 77 96
Mario: 71 90
Luigi: 70 89
Ness: 68 87
Pikachu: 63 82
Kirby: 59 78
Jigglypuff: 48 67

Fall Speed Tests
These tests involve only falling speed. Capture w/ Mupen video capture, look at frames w/ CamStudio player and left/right arrow keys, and see the number of frames where they are in a falling state. Do not count landing, or the step off.

Midheight: Walk off Hyrule green tower.

Falcon: 38
Fox/Link: 39
DK/Yoshi: 43
Ness/Pikachu: 45
Kirby: 51
Mario: 52
Luigi: 55
Samus: 56
Jigglypuff: 59

HighHeight: Walk of Hyrule third platform
Falcon: 51
Link: 52
Fox: 53
DK/Yoshi: 58
Ness: 60
Pikachu: 61
Kirby: 68
Mario: 72
Luigi: 75
Samus: 76
Jigglypuff: 82

Weight Tests
These were done using the Mushroom Kingdom Balance Scale. From heaviest, to lightest:

DK
Samus
Yoshi
Falcon/Link
Mario/Fox/Luigi
Ness
Pika
Kirby
Jigglypuff

These were the tests I did.
Mario = Fox = Luigi
Falcon = Link
DK > Link, Samus
Samus > Mario, Falcon, Yoshi
Link > Mario, Fox
Falcon > Mario
Mario > Ness
Ness > Kirby, Pika
Yoshi > Ness, Mario, Falcon
Kirby > Jiggs
Pika > Kirby

The following were done by putting someone on the left area of Hyrule, next to the wall, and then using Mario’s backthrow [16%], and seeing what percent they die at. Where I stand when the throw is done makes a small % difference, so I ran through the enemy and grabbed as close to the wall as possible. Give these a +1% margin of error. I didn’t give any equal from the previous list a break, I actually tested them. These numbers are before the bthrow.

DK: 155%
Yoshi: 138%
Samus: 136%
Link: 132%
Captain Falcon: 130%
Mario/Luigi/Fox: 126%
Ness: 114%
Pikachu: 107%
Kirby: 104%
Jigglypuff: 93%

I assume some % differences come from falling speed. Those that differed from the weight scale received much testing, so I don't think the margin applies here.

Crouch Cancelling and missing techs
Opponents lying down take 1/2 damage, rounded. Odd numbers round up. Seem to also have 1/2 knockback, well…a little more. Probably this is due to gravity. Opponents who crouch cancel, whoever, don’t get 1/2 damage, but they have even LESS knockback than those lying down.

Test: Mario Dsmash Pikachu on castle base [17%]. Do a usmash when Pika is standing, lying down, and crouching. Make sure Mario’s moves are NOT degredated beforehand.

Standing:


Unteched [usmash did 10% rather than 19%]:


Crouch Cancelled:



Knockback degredation:
Knockback is either calculated BEFORE the hit is given, or AFTER. Because of this, two distinct tests need to be done, and compare knockback [on Hyrule bricks]. If degredation does not affect knockback, one test will have approximately same height, and the other won’t, and the equal one will help me know how it’s calculated. Otherwise, degredation knockback does indeed exist. Samus was used for these tests…just cause. Mario was the usmasher.

Result: Inconsistency w/ both tests indicate that knockback degredation does exist. No conclusive proof on whether before or after % is used in the calculation [higher height on second test could come from the higher damage either before or after the usmash].

After hit calculate knockback test [target 34% after knockback]
- To test degredation height, drag Samus to bricks. Do two usmashes [one away from bricks to make it easier], take snapshot of heighest point.


- To test undegredation height, bring Samus to 34-19, or 15, using tip of drill and jabs. Then usmash.



Before hit calculate knockback test [target 19% before knockback]
- To keep same benchpoint and test degredation height, drag Samus to bricks. Do two usmashes [one away from bricks to make it easier], take snapshot of heighest point. Same picture applies.


- To test undegredation height, bring Samus to 19, using tip of drill and jabs. Then usmash.
 

Daedatheus

Smash Lord
Joined
Jun 10, 2008
Messages
1,137
Location
Toronto & Kingston, Ontario
Good ****! You told me much of this on MSN but it is very organized, clear and useful here. This will come in handy in some effectively subtle ways, especially for snatching KOs through degradated moves, and I'm going to experiment more with crouch-canceling in this game for the dashattack spammers. Cheers
 

AvengerAngel

Smash Journeyman
Joined
Jun 13, 2008
Messages
449
Location
Italy
Wow, that's an excellent work o_o
I already knew of CC and of move's power degradation, but I didn't knew the knockback was reduced too.
Awesome job for the weight tests too. I had the feeling that Samus was heavier than Fox and Link, lol
 

NixxxoN

Smash Master
Joined
Jun 16, 2008
Messages
3,726
Location
Barcelona
great job man!
I didn't know Samus was 2nd heaviest. And LOL @ jiggly dying at 67%, mario's usmash is such a great move
 

Zantetsu

Smash Master
Joined
Sep 1, 2006
Messages
4,413
Location
Springfield, MO
I'd love to see data on falcon's forward throw -> falcon punch. What characters it works on at what percents would be such great information to know. I know the basics for falcon and some other characters but having the exact percents would help a lot.

This is very nice data btw. I'd love to see more things like this :D

 

NixxxoN

Smash Master
Joined
Jun 16, 2008
Messages
3,726
Location
Barcelona
I'd love to see data on falcon's forward throw -> falcon punch. What characters it works on at what percents would be such great information to know. I know the basics for falcon and some other characters but having the exact percents would help a lot.
Has a pretty high range of %. You can do it before the other has 50% or somewhere near it
 

KoRoBeNiKi

Smash Hero
Writing Team
Joined
Apr 26, 2007
Messages
5,959
Location
Brooklyn, NY
Slippi.gg
KORO#668
Very nice guide, waiting for the falcon punch sections, make sure to put in fast falled and short hop falcon punch's
 

Superstar

Smash Champion
Joined
Feb 9, 2007
Messages
2,351
Location
Miami, Florida
I'll test fthrow->Falcon Punch...as soon as someone tells me how to do the thing. XD I'm seconding Falcon now, so PLEASE teach me how to do it. :D Or I might practice, if I find time. College finals end Wednesday, and I really have been putting stuff off for this.

Also, I didn't put this, but it seems to be really weird [about Falcon Punch]. For a while, I was planning NOT to use it for the degredation tests because it was so inconsistent. If you stand right next to the enemy, you deal 25%. A little bit back, but in a specific spot, does 26%. Bit further back does 24%. Also, Falcon Punch is the lowest damage dealing move that requires 4 attacks to regenerate. Only other is Ness's PKT2...Samus's charge shot does not degredate AT ALL, 30% each time. Jiggs Rest and Kirby's Stone are 20%, which is the closest any other move can get.

Also, I figured out Knockback degredation in the off the top tests [which is meant to be a way to quantitize both weight and floatiness in one package]. I accidentely usmashed to bring up damage, then did it again. At the % it was supposed to KO, it didn't even come close [didn't enter the bubble]. So, afterwards, I did some tests to figure it out.

The 1/2 damage came when I was gathering move %. I said some moves get 1/2 damage on opponents lying down. Then I noticed too much of that, and tested it. Also, don't take the backthrow tests as proof, the mushroom kingdom balance is better, it only helps to see "by how much".

I was thinking of doing initial dash length, duration, and speed next. I figured out a way to do it in the bath. First, I count the # of frames for it to end. Then I count the number of dashes to get from one end of Hyrule main floor to the next [and the last portion, count the frames]. Then multiply the number of dashes by the duration to get the speed [or rather, the slowness]. But later, if I feel like procrastinating more or after Wednesday.

Also, sex kicks staling have no formula. If I remember, both Mario and Fox's nairs do 14%, but Mario's weakens to 11% near the end of the move [AND degrades to that XD], and Fox's weakens to 9% [but degrades to 11%].
 

ant-d

Smash Lord
Joined
Jun 3, 2005
Messages
1,314
Location
London, England
I'm not sure what you mean by fall speed. You may mean fall-acceleration or maximum fall speed.

It makes a difference because Link has greatest maximum fall speed. Where as Fox has the fastest fall-acceleration.
Of course it must be difficult for you to test this without the proper software etc.
 

Superstar

Smash Champion
Joined
Feb 9, 2007
Messages
2,351
Location
Miami, Florida
Yar, I was thinking about that. Which is why I did some ratios in division.

Jiggs/Falcon =
high: 82/51 - 1.608
mid: 59/38 - 1.552

But there was nothing substantial in this [I forgot my Physics and Calc, bleh :p Taking Phyics I w/ Calc and Calc II next January]. Then I forgot I ever thought that. XD

It's closer to maximum fall speed than acceleration, and yes, you're right, its hella difficult to test. :p Might as well recheck the recordings I have, still difficult though. If I knew a way to keep the camera zoomed out, I would [hopes training mode has that option EDIT: NO].

If I can increase the height from dropoff to drop point, then I can get much closer to max fall speed [order only though]. Maybe if I DI to the lowest area [I'm pretty sure when it comes to horizontal air movement, it does not affect the falling speed]. And I use a smaller area, I'd get closer to acceleration [still though, margin of error too high].
 

NixxxoN

Smash Master
Joined
Jun 16, 2008
Messages
3,726
Location
Barcelona
I'll test fthrow->Falcon Punch...as soon as someone tells me how to do the thing. XD I'm seconding Falcon now, so PLEASE teach me how to do it. :D Or I might practice, if I find time. College finals end Wednesday, and I really have been putting stuff off for this.
throw with relatively high % (try 50 or 60 or similar) and inmediately jump properly to connect him a falcon punch. I think it also works with high % and double jump. The best falcon players should know it well
 

Superstar

Smash Champion
Joined
Feb 9, 2007
Messages
2,351
Location
Miami, Florida
Hmm, I was thinking a bit more, me knows kinda how to figure them out [what AntD said].

I do tests from the same place [3rd platform], but I time my fall to land next to the **** tent [meaning head right]. I then look at the total time in air, and I look what portion was before passing the second platform, and what portion is after. Using the portion after, I use that for max speed tests [acceleration might have finished by then, and if it didn't, I don't think what's left is statistically significant]. Won't be exact speed differences, but might help get an order. Portion before I use to help with acceleration tests [I know it's not perfect, but I'll manage, I'll do more tests to help me get a better idea].

I did a small pre-test, if I hold right no character will land on the second platform, means once I have time it'll be go.

And thanks Nixxon, I'll try it.
 
Top Bottom