Not a member of Pastebin yet?
Sign Up,
it unlocks many cool features!
- [15:37:55] <&Marty> GreatSage, I've done some tests, let's move the discussion here
- [15:38:36] <&Marty> I had two Deoxys-S, one named Fast with 504 Speed and one named Slow with 503 Speed
- [15:38:57] <%Joim> lol
- [15:39:27] <&Marty> after three Agility each and in Trick Room, Slow moved first consistently throughout two full Trick Room durations
- [15:40:09] <&Marty> I put up Rain Dance (the Deoxys have Swift Swim) and Slow continued moving first consistently in Trick Room
- [15:40:53] <%Joim> so the bug thing is untrue?
- [15:41:10] <&Marty> also note that the two opposing Level 1 Magikarp moved after both Slow and Fast
- [15:41:47] <&GreatSage> hm
- [15:42:12] <&GreatSage> did you check whether the unusual behavior with a pokemon of a "normal" speed, like 300, still existed in bw2?
- [15:42:29] <&GreatSage> (i mean, the basic trick room "bypass" thing)
- [15:42:32] <&Marty> this was done in White 2, yeah
- [15:42:44] <&GreatSage> ok
- [15:43:11] <&Marty> I'm not really sure what to make of these results
- [15:43:28] <&GreatSage> hm, so it's definitely not a speed tie as UPC suggested, right?
- [15:43:40] <&GreatSage> well, by "definitely" i mean "very probably not"
- [15:43:56] <&Marty> right, there must be some other factor
- [15:45:09] <&Marty> perhaps there is a second Speed check if the stat is 0
- 02[15:45:40] * +Jirachee ([email protected]) Quit (Quit: stealth_duck v2.05)
- [15:46:22] <&GreatSage> i am still really confused as to why 1808/1809 would be the breakpoint
- [15:46:27] <&GreatSage> did game freak just use some arbitrary
- [15:46:33] <&GreatSage> "large" number
- [15:46:40] <&GreatSage> that they thought nobody would ever bother to exceed?
- [15:47:26] <&Marty> if that's the case they forgot they invented Speed Boost Ninjask :/
- [15:48:03] <~Cathy> well
- [15:48:10] <~Cathy> xfr has already implemented an extremely accurate simulator
- [15:48:18] <~Cathy> why don't we just take a look at how it handles trick room
- [15:48:30] <~Cathy> http://vsplayer.pokecheck.org/?src
- [15:48:43] <~Cathy> ah
- [15:48:45] <~Cathy> it doesn't need to handle it
- [15:48:49] <~Cathy> because the turn order is already determined for its purposes
- [15:48:57] <~Cathy> well, xfr probably can give us an explanation of the mechanic anyway
- [15:49:00] <~Cathy> when he's active
- 03[15:49:09] * ChanServ sets mode: +v xfr
- [15:49:26] <+xfr> I'm here
- [15:49:36] <~Cathy> hi
- [15:49:42] <~Cathy> do you know how trick room works in gen 5 games?
- [15:50:07] <+xfr> hi, let me see
- 03[15:51:43] * Jirachee ([email protected]) has joined #showdown
- 03[15:51:44] * ChanServ sets mode: +v Jirachee
- 03[15:56:17] * Hyperbeem ([email protected]) has joined #showdown
- 03[16:04:27] * elevator_afk ([email protected]) has joined #showdown
- 02[16:04:28] * elevator_afk ([email protected]) Quit (Client exited)
- 03[16:04:47] * elevator_afk ([email protected]) has joined #showdown
- 03[16:04:48] * ChanServ sets mode: +v elevator_afk
- [16:04:48] <+xfr> ok
- [16:04:55] <+xfr> speed room has a very simple effect
- 02[16:05:28] * %Broolucks ([email protected]) Quit (Ping timeout: 187 seconds)
- 02[16:05:44] * +elevator_music ([email protected]) Quit (NickServ (GHOST command used by elevator_afk))
- 03[16:05:47] * elevator_afk is now known as elevator_music
- [16:05:55] <+xfr> in the comparison value for turn action sorting, the speed is replaced with 0x710-speed
- [16:06:31] <+xfr> so if your speed is over 1808 it overflows
- [16:06:38] <~Cathy> cool
- [16:06:45] <~Cathy> thanks a lot for that information
- [16:07:14] <+xfr> the turn is determined by applying insertion sort on these turn action values
- [16:07:50] <Sakura> I'm trying to set up a custom avatar, I've written it into config.js and put the picture into the folder, How do I use it or assign it?
- [16:08:07] <~Cathy> just for avoiding any doubt -- what does it overflow to?
- [16:08:14] <~Cathy> what is 0x710-(0x710+1)
- [16:08:15] <+xfr> the 0x10000 value depends on action type, 0x1000 is priority, then subpriority (like lagging tail or that berry), then the actual speed
- [16:08:48] <+xfr> it will overflow to 1 apparently
- [16:10:16] <~Cathy> that system is actually pretty clever
- [16:10:23] <~Cathy> it's more optimised than our series of comparisons
- 02[16:11:17] * +Eggy ([email protected]) Quit (Quit: http://www.mibbit.com ajax IRC Client)
- [16:13:49] <+xfr> to be precise the exact calculation is 0x2710-speed but possibly due to an error
- 03[16:13:58] * iss ([email protected]) has joined #showdown
- 03[16:13:59] * ChanServ sets mode: +v iss
- [16:14:12] <+xfr> the precision is truncated to 12 bits for the speed action sorting
- [16:15:18] <+xfr> however, the real speed value is in fact 0x2710-speed because it would overflow the sub-priority value otherwise
- [16:16:06] <+xfr> even without trick room speed precision is only up to 4095
- [16:16:27] <~Cathy> interesting
- [16:16:36] <~Cathy> is there any way to overflow that other than via trick room
- [16:16:38] <~Cathy> ?
- [16:17:31] <+xfr> I don't think so
- [16:17:46] <+xfr> heh actually...
- [16:18:07] <&Marty> oh, that explains why the 2012 Speed Deoxys still 'outsped' two Pokemon with no more than 6 Speed in Trick Room
- [16:18:10] <+xfr> I think with tailwind and scarf Ninjask might hit 4140 at +6
- [16:19:06] <&Marty> that's very interesting
- [16:20:00] <&Marty> thanks xfr
- [16:20:33] <&Marty> I'll go tack on Tailwind to this Swift Swim Agility Deoxys
- [16:21:33] <+xfr> once again the precision problem only applies to the action order decision
- [16:21:52] <+xfr> the end of turn effects also are speed sorted, but they do not have priority or subpriority and thus use the full precision
- [16:22:14] <+xfr> so you if you have an overflow the speed will still be correct for the end of turn effects
- 02[16:23:12] * Hyperbeem ([email protected]) Quit (Ping timeout)
- 03[16:23:23] * V4 ([email protected]) has joined #showdown
- 03[16:23:26] * ChanServ sets mode: +o V4
- [16:24:42] <+xfr> ok correction the precision is actually 13 bits
- [16:24:55] <+xfr> so without trick room the overflow is at 8191
- [16:25:09] <+xfr> still possible to hit with swift speed I think
- [16:26:03] <+xfr> Game Freak mistake was to use 0x2710-speed when they should have used 0x1FFF-speed
- [16:26:06] <~Cathy> so is the effective speed in trick room (0x2710 - speed) & 8191
- [16:26:20] <+xfr> yes exactly
- [16:26:55] <+xfr> &8191 is at 021BC7E6 for the action sort
- [16:27:09] <+xfr> and 0x2710-speed is at 21BCA0C
- [16:27:31] <+xfr> it simply does LDR R0, =0x2710, SUBS R4, R0, R4
- 02[16:28:21] * Snowflakes ([email protected]) Quit (Connection reset by peer)
- 03[16:28:34] * Snowflakes ([email protected]) has joined #showdown
- [16:30:05] <~Cathy> mm
- 03[16:30:16] * prem ([email protected]) has joined #showdown
- 03[16:30:19] * ChanServ sets mode: +h prem
- [16:30:21] <~Cathy> interesting mechanics
- 02[16:30:47] * Menace13 ([email protected]) Quit (Quit: Menace13)
- [16:31:29] <&Marty> hmm, I can hit 12096 Speed, let's see how this goes
- 03[16:32:05] * Menace13 ([email protected]) has joined #showdown
- 02[16:33:12] * %prem ([email protected]) Quit (Ping timeout)
- [16:33:24] <+xfr> 12096 will be treated as 3904, try with 8192
- [16:33:35] <+xfr> and you'll move last against anything
Advertisement
Add Comment
Please, Sign In to add comment