øAslickproductions.org/forum/index.php?PHPSESSID=5f0fck550j2m4m2fpbtkj2vkm1&action=profile;u=25;area=showposts;start=105e:/My Web Sites/Slick Productions - FFIV Message Board/slickproductions.org/forum/index98c5.htmlslickproductions.org/forum/index.php?PHPSESSID=5f0fck550j2m4m2fpbtkj2vkm1&action=profile;area=showposts;u=25e:/My Web Sites/Slick Productions - FFIV Message Board/slickproductions.org/forum/index98c5.html.zxA›h^ÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÈà•¤JîOKtext/htmlISO-8859-1gzip0|ÖJîÿÿÿÿÿÿÿÿWed, 11 Mar 2020 08:19:38 GMT0ó°° ®0®P®€§²ð®A›h^ÿÿÿÿÿÿÿÿó(Jî 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

106
Game Modification Station / Re: Quick Megafix topic
« on: January 22, 2017, 01:20:21 AM »
:noooo:

i had the unfortunate timing of seeing your initial reply when editing my first post, so i didn't know there was trouble until checking the topic by chance 1.5 hours ago.

thanks for testing with Palidor.  oddly, my 10-Speed Gogo (hacked that way months/years ago) does eventually land, but doesn't when hacked up to 40-something Speed.  so it's like the other two characters being further along when Quick is cast is apparently working against them.  also, i am able to get the problem even with Quick+running.

i'm unable to get it with multiple individual, rapid-fire Jumps in place of Palidor, but am unsure whether this makes the issue specific to the Esper, or if i just haven't duplicated the timing involved.  if it is dependent on the Summon, that means we have Palidor complexity stacked on top of Quick complexity to unravel.  :eek:

107
Game Modification Station / Re: Sprite Priority Bug?
« on: January 21, 2017, 06:42:34 PM »
Lenophis: huh, i might've heard of that limit before, but had forgotten about it, and never would've guessed it applied here anyway.  it's 32 sprites per scanline, and 34 8x8 sprite tiles per scanline.  sure enough, checking the Sprite Status in Geiger during that battle with Vomammoths indicates a ton of them with a starting Y coordinate in the high double digits to low 100s.

and from what i read, even the transparent parts of sprites count towards this limit.  i believe this explains what i'm seeing with Terra's damage numerals taking a tiny chunk out of the bottom Vomammoth.  the bottom of the visible digits is above the top of the foe -- but the numerals sprite must extend further down (as sprites in the battle are either 16x16 or 32x32 size).

why didn't Square just give the damage digits a wicked high sprite number (i.e. to lower the priority)?  do they need to take precedence over anything besides the entity receiving the damage?  (and if it's really short, maybe any entity right above/behind it.)  now, this wouldn't get rid of problems like those from the Heal Force sparkle, but seems like one of the simpler measures.

Quote from: TheNattak
The invisible mould is clipping into the sprite it seems: http://imgur.com/a/bcjL4
making it so the moulds do not intersect does indeed fix it from happening, but this is not a feasible fix since some of the formations look bad moving the monster too far up or down.

maybe not in some battles, but certainly a perfect one here.  it's funny to see that FF3usME pic and speculate about some Square employee turning down a layout of bordering moulds, and insisting that "no, no", the small (but eventually devastating) overlap is "just right".

also, condensing these formation layouts to fit on the GBA screen could've wreaked havoc, but it looks like the per-line limit works differently: http://problemkaputt.de/gbatek.htm

108
General Discussion / Re: NFL season 2016- 2017 week 18 (playoffs?!)
« on: January 20, 2017, 10:33:20 PM »
Lenophis: are you gonna update (or reiterate) your AFC pick?  you had Patriots over Chiefs, which is a bit invalid now.

---------

Divisional Round results:

Lenophis = 3-1
Deathlike2 = 3-1
assassin = 3-1

so far:

Lenophis = 6-2
Deathlike2 = 6-2
assassin = 7-1

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

two very close games last week.  in the first, Rodgers was masterful, though young Prescott and Elliott were impressively unfazed by their first playoff game.  a great watch.

for the night match-up, the Chiefs' defense, Le'Veon Bell, and the Chiefs' offense all conspired to keep the Chiefs' D on the field as long as possible, and to make me fall asleep sitting up at the end of the half.  they succeeded on both counts.  "bend but don't break" amazingly continued, but boy, did they bend a lot.  they gave up a crapload of yards, and helped Boswell set a Steelers' record.  maybe if i'd woken up a couple hours later in the morning, or the game order hadn't been switched, i would've been more entertained by this tight game.  it's odd how little Kansas City rushed the ball; they averaged over 4 yards per carry, so it's not like it was futile.  they instead went pass-heavy, and barely converted third downs.

109
Game Modification Station / Re: Sprite Priority Bug?
« on: January 20, 2017, 06:51:49 PM »
it's fun to watch this unfold in an emulator that lets you toggle all the graphics layers.  the game goes nuts moving the sprites between layers.  (i used bsnes v0.073 (Qt GUI) this time, considerably newer than what i recommended before.  the lack of an official changelog makes it hard to pin down when this feature went away, or maybe it's still in the source code waiting to be activated.)

for the monster attacks in that Mammoth battle, the timing suggests that damage numerals are responsible for the missing chunk.  as for Heal Force, maybe it's the top part of the green sparkle?  for me, the chunks have only been temporary; the only time i got them permanently was when being annihilated and letting the battle sit.

and a good theory about the ZSNES shortfall.  i'll have to doublecheck whether the ancient Snes9x 1.39 i use has the same limitation.

110
Game Modification Station / Re: Sprite Priority Bug?
« on: January 15, 2017, 11:26:37 PM »
i don't recall seeing it on the SNES, either.. but i might not have even seen it on vanilla emulated (generally Snes9x 1.39).

it does seem unlikely that ZSNES would dodge a straight-up storage limitation, since i doubt it's nice enough to create extra space for sprites.  which suggests more of a speed or timing issue...  wish i knew more.

111
Game Modification Station / Re: Sprite Priority Bug?
« on: January 15, 2017, 10:46:56 PM »
check this thread:
http://www.ff6hacking.com/forums/showthread.php?tid=2552

Madsiur talks about the limitations.

and this one, namely the second page / last post:
http://www.ff6hacking.com/forums/showthread.php?tid=2563

Stray Aspik has a good theory there.  Magitek Armor is definitely key in this.

finally, this is good reading, but doesn't get into those specific issues:
http://mnrogar.slickproductions.org/phpBB3/viewtopic.php?f=3&t=470

it's more about making sure custom moulds comply with limits.

Quote
I see it happen with 2 large sprites like behemoths when they cast Meteo, and Atma as well, so yeah.

did you get the issues with them on vanilla?

112
General Discussion / Re: NFL season 2016- 2017 week 18 (playoffs?!)
« on: January 15, 2017, 04:15:03 PM »
Pittsburgh, 21-20.

i was on the fence with this game, as well as today's NFC match.  reasoning on the second pick: the schedule strength was similar, while the Chiefs defeated somewhat better teams on average, and both teams were 6 for 10 (including playoffs) against 8-8 or higher teams.  Pittsburgh won big head-to-head, and there were a mere 2 common opponents to possibly outweigh this, but the teams performed similarly here.  and unlike some other playoff match-ups so far, the Chiefs aren't markedly better at home (6-2) than the Steelers are on the road (5-3).  KC and Pitt are very close on both points scored and allowed, but Pittsburgh is better in passing, rushing, and total yardage on both sides of the ball.  their team and QB have more experience and success in the playoffs.

one thing giving me pause is that the Chiefs kept their losses pretty narrow, outside of the one.  the Steelers often lost bigger.  i like their current streak, though.

113
General Discussion / Re: NFL season 2016- 2017 week 18 (playoffs?!)
« on: January 15, 2017, 09:36:04 AM »
i damn near nailed that score.  almost makes up for being grievously wrong on all the rest. ;)  much of it turned out to be an interesting game.  unfortunately, i was out grocery shopping, and missed the better (first) half.

as for today's first game, Cowboys, 27-24

114
Game Modification Station / Re: New bank C3 disassembly, and related docs
« on: January 15, 2017, 05:14:47 AM »
if you apply my Alphabetical Rage patch, then apply the "main1" version of the Rage Checklist Patch, and do steps #1 and 3 from the previous post for Pugs support, this oughta work to restore alphabetization:
http://assassin17.brinkster.net/patches/rage-colors/ff6-try-combine-colored-rage-alphabetical-c3-3.txt

(patch user will need to fill in the two "org" values.  adapt to far jump and/or call if out of Bank C3 space.)

possible optimizations:
- in my BuildLoop, change "BIT $1D2C,X / ... / LDA #$FF" to "AND $1D2C,X / ... / DEC".  or "BEQ / BRA" can be merged into "BNE", unless i feel the need to put something like "TDC" in place of the dropped TYA.  or combine the two changes, and do "EOR #$FF" or "DEC" right after the AND.  yes, it's a bit ugly. :P
- you have 8 bytes to spare before C3/5452, and my extra_rgcol function is a little bigger than your unlk_rg, so they should probably be switched to lower free space use (provided the former is in Bank C3).
- in your rgcolor:
Code: [Select]
       TXA             ; Enemy number
       inc
       beq .skip       ; make Pugs's tbl value 0     <- Table value is defined in "mark6x1" function above
       dec             ; restore enemy number if not Pugs
.skip  CMP $7EAA8D,X   ; Roams Veldt now?

115
General Discussion / Re: NFL season 2016- 2017 week 18 (playoffs?!)
« on: January 14, 2017, 04:30:39 PM »
as for today's other game, there have been no disasters in the news as of yet, so i'll go ahead with New England, 33-16.

116
General Discussion / Re: NFL season 2016- 2017 week 18 (playoffs?!)
« on: January 14, 2017, 03:37:32 PM »
Falcons, 29-26

Seahawks narrowly won head-to-head, and maybe we're supposed to pick the defensive powerhouse over the offensive one in playoffs.  but Falcons had a slightly better record despite a tougher schedule, comparisons against common opponents (and there were a ton!) almost always favor Atlanta, Falcons never lost by more than 9 points (but Seahawks were good in this area, aside from Green Bay blow-out), and Seattle isn't a great road team.

this board's clock is 1.75 minutes fast, btw.

117
Game Modification Station / Re: New bank C3 disassembly, and related docs
« on: January 14, 2017, 02:16:33 AM »
Quote from: me
displaying Pugs there is pretty easy; just put any value other than 255 in the menu slot, iirc.  sadly, that won't work in my Alphabetical Rage patch, where the slot content actually matters.  (the patches conflict anyway; it's just something i recall from development.)  EDIT: hmm, maybe i could work around that, but it's probably not worth the effort or expansion of C3/5418 i'd need.

it could have its uses...  steps needed:

1) data: insert Pugs' FFh into alphabetical order -based list of monster IDs, bringing it from 255 entries to 256.

2) Bank C3:
a. C3/53DB: load some non-FFh value into A.  options include changing the TYA to TXA or TDC, or just dropping it, as A is already 2 ^ n .
b. C3/5433: bring back Square's "LDA $E5".  they were ahead of their time!  then do "TAX / LDA $C2FAD0,X".  (the Skills list doesn't compress, so we can say what Rage would be in a given position.)  that'll be 4 bytes shy of fitting (with "CMP #$FF" ==> "INC"), but i can steal Novalia Spirit's Esper name blanking optimization.

3) Bank C2: i had a 255-length alphabetized list largely for this bank's concerns.  we don't want Pugs' ID output there, because we can't use the rage, and it could overflow the $3A9A counter if somehow all others are possessed.  so when going 256-length, we would need "CMP #$FF / BEQ" at about C2/584A to skip Pugs.  C2/585F's "CPX #$FF" can be dropped, but i still need 2 more bytes.  it looks like the SwdTech and Blitz "TDC"s can go, though i'd sort of prefer to keep my changes in the Rage part of the function.  any ideas?

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

also, Novalia: do you have the Rage coloring patch source code differences for the version that blanks unlearned+unmet monsters?

118
General Discussion / Re: NFL season 2016- 2017 week 18 (playoffs?!)
« on: January 13, 2017, 03:01:52 AM »
...maybe the plane crashes into the Patriots' practice facility?  or they charter an emergency flight to receive elite medical help for their boils.

119
General Discussion / Re: NFL season 2016- 2017 week 18 (playoffs?!)
« on: January 12, 2017, 02:03:10 AM »
haha; i was right for 3 quarters, anyway.

Wild Card Round results:

Deathlike2 = 3-1
Lenophis = 3-1
assassin = 4-0

maybe we should start listing catastrophes that would allow Houston to win?  i'll kick things off with plane crash, and plague of boils.

120
General Discussion / Re: NFL season 2016- 2017 week 18 (playoffs?!)
« on: January 08, 2017, 03:39:02 PM »
Packers, 21-18

while my point gaps were pretty close (and wrong) on all games, this is the only one i hesitated on picking the victor.  it could shape up to be the sole interesting game.  having two able-bodied starting quarterbacks is monopolized by NFC contests.