øAslickproductions.org/forum/index.php?PHPSESSID=5f0fck550j2m4m2fpbtkj2vkm1&action=profile;u=25;area=showposts;start=45e:/My Web Sites/Slick Productions - FFIV Message Board/slickproductions.org/forum/index59e9.htmlslickproductions.org/forum/index.php?PHPSESSID=5f0fck550j2m4m2fpbtkj2vkm1&action=profile;area=showposts;u=25e:/My Web Sites/Slick Productions - FFIV Message Board/slickproductions.org/forum/index59e9.html.zx@›h^ÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÈà•¤¶ëOKtext/htmlISO-8859-1gzip8:Ö¶ëÿÿÿÿÿÿÿÿWed, 11 Mar 2020 08:19:36 GMT0ó°° ®0®P®€§²ð®?›h^ÿÿÿÿÿÿÿÿ>'¶ë Show Posts - assassin

Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Messages - assassin

46
General Discussion / Re: Someone explain this Donkey Farm thing
« on: April 28, 2017, 03:36:04 PM »
that's correct.  Amano originally intended Shadow to be a key NPC, the lead ranch hand of the Donkey Farm.  however, it was determined that the cartridge space needed for the braying SPC samples would be prohibitively expensive.  there was also concern that the ESRB would slap a strict rating on the game due to the sheer power of donkey kicks, even when presented as slapstick in nature with the ranchers fully recovering from them.  the 1990s were truly a different era, both in terms of technology and cultural mores.  however, this was one case where compromise didn't hinder the final product.  Amano, while initially disappointed, turned in some of his finest work on Shadow's overhauled garments; his name also fit better once the brightly-colored kerchief and large straw hat were abandoned (his headband retaining elements of the former as a subtle nod to the original sketches).

47
there is no point.

1) but as far as a cartridge player, Square's target audience, knows, the Rage never is learned.
2) for Novalia Spirit or me, enabling the out-of-battle listing is just there for information for the curious.  that's why it's present as an added option to his patch, not a requirement.

making it work is non-trivial:

Quote from: Novalia Spirit
Getting this to work in battle would probably not be too difficult either. I'm guessing I could just assign Guard's value (#$00) to an unlocked Pugs Rage, and then add a check to see if the Rage is in the very first Rage slot. This would cause a problem with the variable holding the current Rage ($33A8,X), but this could probably be circumvented by setting an unused flag and adding the necessary checks.

and there are still the issues discussed in rage-list-alpha.txt .

48
this is covered in my patch's rage-list-alpha.txt .  there isn't a single post in this thread by either author suggesting any patch involved can make Pugs usable in battle, so your surprise at this surprises me. :P

in an 8-bit menuing system, some 0-255 value has to represent null.

49
Game Modification Station / Re: Fancy Walking patch issue
« on: April 05, 2017, 04:20:05 PM »
thank you for your research and diagnostics on this, and Xenovant for his initial debugging.  it will prove useful/central in updating the patch.

Quote
If we still want the ability to talk while moving, we'll have to add a "talking to an NPC" check before the movement.

besides taking space, are there any suspected side effects or issues with this?

50
Game Modification Station / Re: Remaining bugs in FF3us
« on: April 05, 2017, 01:20:31 AM »
probably "Setzer resents bird-hating comrades" in Master ZED's Bugs Guide.

51
yeah, i was meaning to enlist him for that, after seeing his C0 work with the party switching bugs, and the needed adjustment to occupy two tiles at once.  is there something more inherently complex about diagonal walking where adding cardinal walking's function calls is a no-no, short of a big overhaul?  or is the remedy something simple?

52
seeing as how the patch behavior isn't technically incorrect, and the issue resolves after the needed Veldt battle, i'd tend to say that making it telegraph the unlocking that'll occur in that battle isn't worth the effort. [EDIT: maybe it is.  regardless, am glad you documented the issue here.]

53
OK, no, apparently it doesn't happen with White Drgn. Now I can only guess that it has something to do with the formation oddities of SrBehemoth.

argh, that was one of my first suspicions, but i wrongly ruled it out, because i can't remember the details and order in which this crap happens.

so when you finish the SrBehemoth 2-part battle (without Vanish + X-Zone scumbaggery, of course): the game marks the First Battle Formation as met for Veldt purposes, but on the basis of the current formation holding a sub-255 enemy ID.  a weird hybrid, yes, but seems intentional given how the game sets an event bit, and later uses it to swap between the two formations at battle start on a Veldt meeting.  (ultimately resulting in you meeting undead SrBehemoth twice per Veldt cycle.)

this nonsense will have to be simulated in some way by the patch.  there's hardcoding (e.g. enable monster #127 whenever formation #452 has been met, or slightly more flexible, enable #127 whenever a formation containing #281 has been met); this route is shorter and easier, but lame.  then there's using the Formation Interchange data and the event bit to follow how the game does it in the first place.  long, but noble.

-------

EDIT: because i am mentally challenged, i was envisioning the latter doing the interchange check within the formation loop.  slower than dirt.  better to have a separate loop that just goes through all 8 Formation Interchange entries.  if an activated first-halfer is found, then convert the second-halfer's formation # into Byte,Bit (a C2/5217 far-call helper would be nice), and call "mark6x8" for it.  EDIT2: and i guess you could double-check first that the first-halfer formation was actually met, if you don't want to rely too much on the event bit.

54
1)
does it still happen without my stuff applied?

2) random thought: Vanish + X-Zone is for scumbags.

------------

EDIT:
Quote
When the patch sees SrBehemoth's MSB as a 1, it neglects to add it to the Rage list.

the living one (#281), sure.  that doesn't explain why undead (#127, i.e. the only Rageable one) wouldn't get added.

55
Game Modification Station / Re: Kefka kamikaze
« on: March 16, 2017, 01:33:15 PM »
Quote
I guess a more realistic fix would be to add a condition that there's someone left alive in the entire 12-character line-up, but that would be pretty damn tricky.

i'm favoring this way, too.  if backup characters can step into a multi-part battle, why should they be incapable of standing over Kefka's disintegrated remains and nodding so the ending can commence?

can the game ever have dead characters in the $0205 list, or does it pre-screen for that, or pre-revive them even?  i don't recall how that works.

honestly, however annoying Kefka's long death animation may be if you didn't truly win, i don't find it erroneous -- he did die, after all.  all i find outright wrong in the video is the long wait after that, and the WoB music.

the tricky thing is, while the monster script is a good place to "get in front of" the special battle event and animation, its language is limited in flexibility for doing what one wants (unless you write a custom command).

56
did you do both #1 and #3 here:

http://slickproductions.org/forum/index.php?topic=2245.msg26820#msg26820

?  in #3, when i say "going 256-length" and "C2/585F's 'CPX #$FF' can be dropped", that means expanding the C2 loop to 256 iterations.

57
if Diddy stands in the vertical shaft, just clear of the ceiling, with the barrel held in front of him, can it be crushed without him?

58
interesting!  i wonder whether her jumping ability opens up new exploits.

as for why she couldn't do it the other way, maybe her barrel put down (or recovery therefrom) is slower, or has a different trajectory.  (to be sure, the former is different from what i was considering in the last sentence of my prior post; that was about having to wait until the barrel hits a wall tile border, in the event Dixie's "put down" routine lacked a breakage check.)  that could give time for both the breakage to be triggered and the barrel to lose its "is on ground" value, assuming the former is consistently done first.

can barrels be ceiling-crushed while being held by a character, or must they be on the ground?  doubtful it would trigger the bug; i'm just trying to pin down the breakage requirements.

59
another thing (i can't edit that post no more!): if there is the wall acting as a ceiling deal, is the crush of a barrel against/within an impassable tile triggered right away, or does it wait until the top of the barrel touches the bottom of a tile border?  and if there is, as you said, a " 'check for barrel breakage' function [in] Diddy's 'drop barrel' function", is it subject to the same alignment wait?

playing while knowing where each wall tile starts (grid lines would be nice) and using frame advance, i wonder whether Dixie putting down a barrel would have slightly more wait until breakage than Diddy.

60
i'll take a leap of faith that DKC1 has some relevancy here, as i find more info on that, the games all look the same, and dunno where i put my DKC2 ROM.  iow, airtight reasoning. ;P

- http://www.dkc-atlas.com/forum/viewtopic.php?f=38&t=1199 , see "Public DKC RAM Table".  testing on Ropey Rampage, the first DK barrel has a value of 26h in the 7E:0D45 table, but then becomes 0 when destroyed.
- i did some cheat searches trying to pin down changes in this DK barrel's states.  contenders:

http://assassin17.brinkster.net/dkc1/dk-barrel-high-ropey-rampage-RAM-addresses-narrowed.txt

[EDIT1: listed values are from testing with Donkey.]

as the one note hints, there might be some variability as to what address a given barrel uses, depending on the path you take to reaching/grabbing it.

i didn't find an outright "Barrel Breaking" flag or constant among those values, but they could be useful if they (or something similar) are used in DKC2.  a fix might be a matter of testing certain ones, and/or making sure they're consistent with each other.  or it might involve altering one or more.  suppose, for example, that a "barrel is on ground" value is what lets you pick up the breaking barrel here, and why i've never heard about picking up a barrel you've just chucked at (or walked into) a wall.  in that case, finding the code that triggers the breakage and inserting a change of said value to something else might remedy things.  (EDIT2: like you said, "properly implement the 'can't pick it up anymore' property in time".)

that's an alternate path to stopping the barrel breakage in the first place, and not one i'm necessarily saying is better, but either way can achieve some needed consistency.

EDIT2: break-pointing those cheat-searched values was able to get me in the vicinity of the DKC1 "pick up barrel" and "put down barrel" functions.

-----

EDIT2: having a barrel somehow be crushed "between" a floor and a vertical wall is a bit dumb, but maybe it'd be too involved to change.  if you put down the barrel in a proper horizontal position (i.e. corresponding to old wall edge) while in a clearing where the walls are wider, then original wall width is reached again, can the stationary barrel be crushed against the side of a wall?  or is the breakage limited to during a Kong's barrel lowering?

also, while going the route of stopping perpendicular breakage makes sense, could Diddy still quick-pick up a barrel that's crushed against the ceiling of a clearing with some precise timing?

i'm guessing that a wall is seen as a "ceiling" in that the barrel, slightly overlapping the wall, is trying to enter an impassable tile from below..  but why does this overlap exist in the first place: graphical presentation?  making sure barrel-toting Diddy needs to walk _into_ a wall to open a secret passage?