There is probably a list for each hitbox that tells what slots have been hit. This list can only carry 12 entries.
I tried finding this in memory but I'm not good at using Cheat Engine and it takes minutes to make one search...
I'm not an expert in CE, Dolphin, or SSBM but I do enjoy the puzzle of finding stuff like this out.
I'm going to make a few assumptions here.
1. You can create a hitbox that will last indefinitely.
2. When you load a state with Dolphin it puts everything back in the same memory location.
3. That the hitbox list is some sort of linked list that gets appended to, and existing entries aren't 'pushed back' by new entries.
So, try the following. Create a hitbox and save state.
Get something hit by the hitbox. Do a CE search for unknown value.
Load state, get something different hit by the hitbox. Search previous results for changed value.
Run around a bit, and do a few narrowing searches for unchanged value.
Load state, run around and do different things for a few seconds, then get the same thing hit by the same hitbox and do another narrowing search for unchanged value.
Hit 10 more things (or whatever the max amount is before rolling over the list) and do an unchanged value search after each one.
Hit 1 more thing, and do a search for changed value.
That should get the list of results pretty narrowed down.
After you're done the first two searches in CE (and any time you're searching for anything) the speed of the searching should improve drastically as results are removed. If 'every' search takes minutes, then you need to improve your searches to get that result list down faster.
http://www.youtube.com/Noxxa
<dwangoAC> This is a TAS (...). Not suitable for all audiences. May cause undesirable side-effects. May contain emulator abuse. Emulator may be abusive. This product contains glitches known to the state of California to cause egg defects.
<Masterjun> I'm just a guy arranging bits in a sequence which could potentially amuse other people looking at these bits
<adelikat> In Oregon Trail, I sacrificed my own family to save time. In Star trek, I killed helpless comrades in escape pods to save time. Here, I kill my allies to save time. I think I need help.
Joined: 8/14/2009
Posts: 4091
Location: The Netherlands
It looks like Bowser SDIs slightly left and down, which might cause him to hit the blast zone sooner than normal. He also doesn't DI/DIs straight down, though I don't know if that's normal. 20XX has customizable DI and SDI settings for the AI, which can affect events like this (I found out about this back when I made a TAS of Trouble Puff).
EDIT: Just compared the 20XX video and a tied WR video frame by frame, it's almost definitely SDI causing the difference.
http://www.youtube.com/Noxxa
<dwangoAC> This is a TAS (...). Not suitable for all audiences. May cause undesirable side-effects. May contain emulator abuse. Emulator may be abusive. This product contains glitches known to the state of California to cause egg defects.
<Masterjun> I'm just a guy arranging bits in a sequence which could potentially amuse other people looking at these bits
<adelikat> In Oregon Trail, I sacrificed my own family to save time. In Star trek, I killed helpless comrades in escape pods to save time. Here, I kill my allies to save time. I think I need help.
Actually, could the Samus BtT TAS record be attributed to "Mario 64 Master" rather than "Scott Rogoff"? I've always shown up in record lists over the years as "Mario 64 Master", so I think that name will be recognizable for more people. Thanks. :)
@Mario 64 Master, I changed it.
I tried to calculate the Btt TAS total here, but I'm bad at it.
I came up with 2:43.30 but I'm not sure it's correct since it's a large margin from the previously listed 2:44.10 (actually 2:44.08 if you add each time).
I used a lookup table from Practical TAS' Peach record history video and went over each character's time.
12+7+3+7+7+6+7+6+4+3+5+11+8+10+7+5+3+4+5+6+6+3+2+7+4 seconds of all characters
=148 sec
41 + 24 + 10 + 37 + 48 + 35 + 55 + 27 + 49 + 16 + 55 + 43 + 57 + 22 + 49 + 28 + 37 + 22 + 41 + 53 + 40 + 37 + 39 + 53 frames (milliseconds of all characters converted to frames with the lookup table)
=918 F
---
918 F = 15 sec and 18 frames
18 F = 0.30 sec
15 sec + 0.30 sec = 15.30 sec
148 sec + 15.30 sec = 163.30 sec = 2:43.30
Thanks for the update!
Your math is correct except you missed one of the frames numbers. Instead of:
41 + 24 + 10 + 37 + 48 + 35 + ...
it should be:
41 + 24 + 10 + 37 + 48 + 48 + 35 + ...
So your total is too low by 48 frames.
I wrote a Google Sheets script and confirmed that the total of scores listed in your table right now is 2:44.10.
Thank you, Mario 64 Master. I corrected it on the records page.
---
I'm looking for Action Replay codes / hacks / mods:
Any one of the following
* Make the distance display in Homerun Contest work for Player 1
* Make it so you can change character for Player 2 and control Player 2 when entering Homerun Contest.
* Enable distance display when entering Homerun Contest from debug menu.
And
* Enable the "How to Play" texture in the background of the Pokémon Stadium stage.
See How to Play video and TCRF.
If anyone can provide help for these, thanks!
I have calculated that the Btt total is now 02:44.06
and the Events total is 07:18.11.
@Mario 64 Master
Can you verify this? If my values are correct I should be able to calculate the time in the future.
Lua https://pastebin.com/9qfhAVG6
----
I added real-time records to the SSBM TAS records page.
While comparing TAS vs. RTA, I found the following which might be interesting to look at:
Bowser TAS 7.62 uses PAL (117 weight),
Bowser RTA 8.15 uses NTSC (118 weight)
Yoshi TAS 6.80 uses NTSC (108 weight),
Yoshi RTA 7.21 uses PAL (110 weight)
I don't know how much weight influences falling speed or landing lag.
I added Events RTA records to the SSBM page.
For HRC I need more information:
1) Where can I find all the current RTA HRC world records?
2) How can I calculate meters <-> feet?
3) How can I calculate the distance total?
---
I have added both console-only and emulator times in the RTA section of the SSBM records page. I think with this I can satisfy both parties, those that want the inclusion of emulated times and those that don't. With this change, I need more information:
4) What is the WR for Btt Samus (on console)?
5) Are any more RTA records emulated and not marked as such?
---
Also
6) Are there any RTA records I missed (especially for Events)?
4) What is the WR for Btt Samus (on console)?
5) Are any more RTA records emulated and not marked as such?
Doc: 13.11 by mudi was done on emulator, console record is 13.31 by sockdude1
Peach: 9.46 by sockdude1 was done on emulator, console record is 9.53 by LDA
Fox: 6.40 by mudi was done on emulator, console record is 6.52 by sockdude1
Samus: 7.75 by Mario 64 Master was done on emulator, console record is 7.87 by LDA
Jigglypuff: 7.33 by mudi was done by emulator, console record is 7.62 by sockdude1
Also, I got 9 targets on G&W riddle on emulator: video
Thank you sockdude1. I'm improving the page every day, step by step.
While looking at regions for each record, I noticed the knee difference for C. Falcon.
The 6.92 TAS by AJP_anton was done on PAL. Does not use knee at all.
The 7.43 RTA by marth1 was done on NTSC. Uses one weak knee.
---
So I tested hitstun for the aerials:
PAL weak knee 3 F
NTSC weak knee 4 F
PAL strong knee 8 F
NTSC strong knee 8 F
PAL Upair 6 F
NTSC Upair 6 F
PAL strong nair first hit 4 F
NTSC strong nair first hit 4 F
PAL weak nair first hit 3 F
NTSC weak nair first hit 3 F
---
I guess Upair is very practicable in TAS as it is quick enough to compensate for the higher hitstun, and weak knee takes too long to come out.
So this knee difference might not lead to a timesave after all.
The knee difference I wrote about in my previous post was already put to good use, 7.41 RTA by samthedigital.
-----
Aside from these interesting BTT differences quoted below
Bowser TAS 7.62 uses PAL (117 weight),
Bowser RTA 8.15 uses NTSC (118 weight)
Yoshi TAS 6.80 uses NTSC (108 weight),
Yoshi RTA 7.21 uses PAL (110 weight)
I also found the following interesting differences worth looking into:
* It seems Fox can run very slightly faster on NTSC (maybe because of his weight difference or an overlooked dash speed difference).
Event 6 TAS uses PAL Fox. Maybe you can save time if you use NTSC Fox. (Not true)
* Falcon has 8 hitstun frames for both NTSC strong knee and PAL strong knee, however NTSC has slightly higher damage/knockback.
Event 8 uses PAL Falcon. Maybe you can save time if you use NTSC Falcon. (Not true)
* Event 14, NTSC Fox has more weight. Use PAL, where Fox has less weight, to knock him out faster?
* Event 16, NTSC Kirbys have less weight. Use NTSC to knock Kirbys out faster?
* Event 20, PAL Fox has less weight so you might save time using PAL version.
* Event 22, PAL Mario has less weight so you might save time using PAL version.
* Event 23, PAL Fox has less weight so you might save time using PAL version.
* Event 26, PAL Fox has less weight but Yoshi has more weight. Maybe you can save time by manipulating different characters.
* Event 31, doesn't have a TAS yet. Consider that PAL Mario has less weight so the PAL version might be better to use.
* Event 36, TAS uses the PAL version.
PAL Kirby has more weight (70→74), but PAL Fox has less weight (75→73).
* Event 37, TAS uses the NTSC version.
NTSC Ganondorf Down Aerial: 22%, 50 knockback
PAL Ganondorf Down Aerial: 21%, 40 knockback
It seems Down Aerial has a late hit. As far as I tested hitstun, I got these results:
NTSC 9 frames first and late hit.
PAL 8 frames first, 7 frames late hit. Both hits deal 21%. The late hit starts pretty much 2-3 frames after the first hitbox is out.
I don't think you can save time by switching to the PAL version, as reducing hit stun on Ganondorf isn't worthwhile. Hit lag on female Wireframe in this event was 14 frames in both NTSC and PAL.
* Event 38, it seems Mario is the last character to die (only by 1-2 frames).
There is no TAS yet, and the RTA uses NTSC.
It might be worth using the PAL version has Mario has less weight.
* Event 46, there is no TAS yet.
PAL Marth has less weight so it may be better to use the PAL version.
* Event 51, I tested for Giga Bowser's weight but there doesn't seem to be a difference between NTSC and PAL.
Joined: 8/14/2009
Posts: 4091
Location: The Netherlands
MUGG wrote:
* It seems Fox can run very slightly faster on NTSC (maybe because of his weight difference or an overlooked dash speed difference).
Event 6 TAS uses PAL Fox. Maybe you can save time if you use NTSC Fox.
* Falcon seems to have 8 hitstun frames for both NTSC strong knee and PAL strong knee, however NTSC has slightly higher damage/knockback.
Event 8 uses PAL Falcon. Maybe you can save time if you use NTSC Falcon.
Any sources for these claims? As someone who's been involved in my local competitive Melee scene which has switched from PAL to NTSC within the past year, I'd be very interested in undocumented/unknown differences between NTSC and PAL outside of what already was known and what was discovered through datamining at the time (which included various changes to e.g. Kirby's hitboxes that weren't known at all until last year)
http://www.youtube.com/Noxxa
<dwangoAC> This is a TAS (...). Not suitable for all audiences. May cause undesirable side-effects. May contain emulator abuse. Emulator may be abusive. This product contains glitches known to the state of California to cause egg defects.
<Masterjun> I'm just a guy arranging bits in a sequence which could potentially amuse other people looking at these bits
<adelikat> In Oregon Trail, I sacrificed my own family to save time. In Star trek, I killed helpless comrades in escape pods to save time. Here, I kill my allies to save time. I think I need help.
@Mario 64 Master, thank you for verifiying.
I think addition of all Meters scores to calculate the total is correct.
@Mothrayas, I did a brief test yesterday, running across Final Destination.
Today I did another test and the running speed seems the same. So my claim doesn't hold true. The difference happened because my Final Destination test was done in PAL 50 Hz mode, this HRC test was done with 60 Hz mode.
I tested Falcon's knee hitstun on a target few days ago, see previous posts.
I tested again with Falcon kneeing a Samus on PAL60 and NTSC and it is 8 frames hitlag for both versions. However, I was not able to save any time over SW-1988's 9.93 TAS.
If you are interested in version differences, you can find a good documentation here. It covers pretty much everything. I contributed what I know to that page. The only thing not yet documented there is Mario's Tornado difference between NTSC 1.0 and other versions, which saved time in Event 1.
Of course there is always the possibility for more stuff that hasn't been found yet.
I think PAL is the most balanced version of Melee for competitive play, but I might be biased.
Joined: 8/14/2009
Posts: 4091
Location: The Netherlands
MUGG wrote:
I tested Falcon's knee hitstun on a target few days ago, see previous posts.
I tested again with Falcon kneeing a Samus on PAL60 and NTSC and it is 8 frames hitlag for both versions. However, I was not able to save any time over SW-1988's 9.93 TAS.
I am aware that strong knee hitlag is the same, my point was about the claim that NTSC strong knee has more damage/knockback. Which according to some quick testing on my end is incorrect (strong knee damage/knockback is identical between NTSC/PAL), and if it were, I'm pretty sure it's something that would have been figured out in the process of PAL -> NTSC patches being made by UnclePunch and others, which is where the other formerly undocumented changes came from.
And yes, I'm well aware of the ssbwiki page about version differences - these differences have been a very hot topic in my community and me and other people have done very extensive research on it, which is why I'm sceptical of these new claims of undiscovered/undocumented changes.
http://www.youtube.com/Noxxa
<dwangoAC> This is a TAS (...). Not suitable for all audiences. May cause undesirable side-effects. May contain emulator abuse. Emulator may be abusive. This product contains glitches known to the state of California to cause egg defects.
<Masterjun> I'm just a guy arranging bits in a sequence which could potentially amuse other people looking at these bits
<adelikat> In Oregon Trail, I sacrificed my own family to save time. In Star trek, I killed helpless comrades in escape pods to save time. Here, I kill my allies to save time. I think I need help.
----
I added real-time records to the SSBM TAS records page.
While comparing TAS vs. RTA, I found the following which might be interesting to look at:
I tested Falcon's knee hitstun on a target few days ago, see previous posts.
I tested again with Falcon kneeing a Samus on PAL60 and NTSC and it is 8 frames hitlag for both versions. However, I was not able to save any time over SW-1988's 9.93 TAS.
If you are interested in version differences, you can find a good documentation here. It covers pretty much everything. I contributed what I know to that page. The only thing not yet documented there is Mario's Tornado difference between NTSC 1.0 and other versions, which saved time in Event 1.
Of course there is always the possibility for more stuff that hasn't been found yet.
I think PAL is the most balanced version of Melee for competitive play, but I might be biased.