øAslickproductions.org/forum/index.php?PHPSESSID=5f0fck550j2m4m2fpbtkj2vkm1&action=profile;u=25;area=showposts;start=195e:/My Web Sites/Slick Productions - FFIV Message Board/slickproductions.org/forum/indexf46c-2.htmlslickproductions.org/forum/index.php?PHPSESSID=5f0fck550j2m4m2fpbtkj2vkm1&action=profile;area=showposts;u=25e:/My Web Sites/Slick Productions - FFIV Message Board/slickproductions.org/forum/indexf46c-2.html.zxD›h^ÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÈ0Pd{OKtext/htmlISO-8859-1gzip0|Ö{ÿÿÿÿÿÿÿÿWed, 11 Mar 2020 08:19:40 GMT0ó°° ®0®P®€§²ð®C›h^ÿÿÿÿÿÿÿÿý1{ 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

196
Final Fantasy V Research & Development / Re: FFV Damage Calculator
« on: May 12, 2016, 09:40:17 AM »
i can't find where that weapon property byte is loaded in the C2 disassembly.  the C2/9AA9 loop does include D1/0005, and C2/9C2E will put an OR'ing of five of those bytes into STA $2039,X for armor.  best put an emulator trace on the specific D1/0005,index of the weapon you're seeking, and use the disasm to follow where it winds up from there.

197
Final Fantasy V Research & Development / Re: FFV Damage Calculator
« on: May 12, 2016, 08:11:39 AM »
which weapon FX?  are you talking about "Magic Sword Drain" and "Magic Sword Psych"?  and is the question in reference to the "Target Can't Evade vs. Attack type" bonus, or just generally speaking?

if the latter, do you mean the different formulas (Damage Types), or add-on spells/etc.?

198
Game Modification Station / Re: Quick Megafix topic
« on: May 12, 2016, 07:59:14 AM »
to recap.. one bug, "Infinite turns w/ Quick and X-Magic" (explained in Master ZED's guide), is in the process of being dealt with.

aided by info in this post:
http://slickproductions.org/forum/index.php?topic=2219.msg25258#msg25258

and discussed in the 10 posts starting here:
http://slickproductions.org/forum/index.php?topic=2219.msg25265#msg25265

the source for the pre-alpha test patch (FF3us) is here:
http://assassin17.brinkster.net/forum-posts/quick-gembox-try-fix1.asm

it's held up in my limited tests, and Xenovant's more involved trials (thanks!).

however, i can't truly say WHY it works, nor that it works all the time or lacks side effects.  a change being mulled is only skipping the relocated code on Quick's final turn.

besides the Function C2/091F edited by the patch, other areas of interest are C2/0168 and C2/01B7.  Variable $3AA0,index almost certainly will be involved in whatever the final fix is.

(Imzogelmo: mnrogar's site also has lots of savestates, though they're for an LLG.  you can always save conventionally on them, then sic Lord J and Jumper's FF3h on the .SRM for stronger characters.)

199
Game Modification Station / Re: WIP Patch: Permanent Morph bugfix
« on: May 12, 2016, 07:17:12 AM »
Sorry magitek, I hit modify instead of reply and totally botched your triple post. Feel free to modify it further and slap me with a wet trout. :isuck:

haha; i did that to you once on Mnrogar's a few years back.  i had to recall your post as best possible (fortunately, it wasn't that long).  there should really be a "Toddler Proof" checkbox that super-users can (un)click before login, given most visits don't require moderating or administrative work (particularly on this forum, in a reality where most spambots haven't played Dynasty Warriors  :happy:)

also, you should have retaliated against ME; what the hell's the matter with you?!?!

.
.

(please don't!  :eek:)

Quote
I was just going to say that you could always do a fix A/fix B solution if there are caveats or multiple interpretations to how it could reasonably be fixed.

magitek: are you still considering the C2/4BB2 relaxation version?  iirc, we agreed it was:
- stronger in terms of minimizing the chance of a Morph user with no "fuel" left continuing to receive the status' benefits (e.g. higher Defense).
- weaker in terms of a certain-timed Stop/whatever spell.  (i'm foggy, so see a few of our prior posts for specifics.)

it's understandable if you've put it off, as juggling both at once would be a bitch to code, and make for confusing discussions.  if you've nixed it, also understandable, though i'm curious as to the reason, given i tended to see the approaches as on roughly equal footing.

200
Game Modification Station / Re: WIP Patch: Permanent Morph bugfix
« on: May 12, 2016, 06:54:48 AM »
I lean towards thinking it makes more sense to pause periodical damage during Freeze, too. But I'm aware that it is a contested topic and as such, strays from pure bugfix territory and will not be including that change in the patch (maybe in the .asm as an optional addon).

yeah, it gets tricky..  for instance, if the periodic damage/healing is coming from equipment, the gear isn't necessarily frozen, but the user they're giving the status to is.

Quote
Quote
    correct me if i'm wrong..  this risk presents once you do the following:

    1) stop clearing $3EE2
    2) Function C2/0B01 handling Haste/Slow toggling
    3) dropping the 2048 decrementor cap


        If I understand the way you described it correctly, zeroing the low byte of $3EE2 ($3F30, as you posted previously) should eliminate this risk? Meanwhile, waiting to zero the supply until we stopped calling C2/0B01 should, uh, eliminate it again (or more).

my point of that 3-item list was that it's hard to consider the C2/0B36 call as culpable for an issue that happens *after* you do three things to modify the game (none of them mandatory for a bugfix, as there are alternate ways, and one of them super-optional).  so the call is not a problem in the game but rather in a very specific reworking thereof.

but yes, agreed on the full timer zeroing being a fix for the loophole i first mentioned.  however, my follow-up issue regarding the Haste/Slow recalculation requires either:
- adding a "BEQ 0B2C_equivalent" before the "JSR $4792" in C2/0B01, and having a nonzero fallback decrementor.  (initially, i said the 2048 cap, but the $3F30 zeroing means even 1 would suffice here.)
- dropping the C2/0B36 call.

Quote
Quote from: assassin
  but a devil's advocate argument to fuck with myself:
            if the Morph timer counts down from 65535 to 500, we don't update the Morph supply.  yet when it goes from 500 to 0, updating Morph supply is now vital?  while 0 is indeed a special case, is it special enough to say that a change of 500 outweighs a change of 65035?

        Not trying to sound facetious, but that kind of sounds more like a devil's strawman. :P

huh?  i was actually challenging my own viewpoint there, by saying: if depleting a whopping 65035 points off the Morph timer isn't meaningful enough (by Square's standards) to update the supply, why should depleting the last mere 500 points -- admittedly a special case -- be enough?

Quote
Can you think of anything concretely screwy that can happen from not zeroing the supply right away, considering Revert will force it's way into the next available action slot? Maybe if autocast got in the way of Revert (Condemned), since those have priority? But, Terra dying doesn't thwart the supply being updated normally, does it?

        I kind of agree it would make more sense to decrement the supply at the same time as decrementing the timer, kind of. But luckily, the supply stays intact for me to use as a base for the decrementor in this patch.

nothing offhand.  C2/4678 oughta catch the Morph removal.  but i don't always want to wait to see the concrete, instead assuming some might be lurking around the corner, or ready to fall from the sky. ;)

yeah, there is a bit of an artificial separation between the two values (but understandable for coding purposes).

if RAM usage weren't a concern, you could always keep the original decrementor, and have an active copy that's potentially status-modified.

201
Game Modification Station / Re: WIP Patch: Permanent Morph bugfix
« on: May 12, 2016, 05:25:11 AM »
That sounds like a good idea, you should add that to the wiki. I noticed some descriptions where your comments add to the understanding of the bit (you have "flag target's ATB gauge to be reset?" and the RAM map has "psyche was just cleared", for example).

thanks, will do.  my $3AA0, Bit 4 description will also be an upgrade to it, but for the most part, i've lagged what they've got there.

Quote
Quote
[re CPX $3EE2 in C2/0B01:] looks like you talked yourself out of it..  but to chime in, i don't favor sacrificing capability just to save a few bytes.

More like I talked myself into it, except I still can't find a way to shrink C2/09D2 enough to fit a jump to C2/0B01.

ah, guess i misinterpreted the final conclusion of the edits in that post.

Quote
1. There's that TYA.
2. I don't think the CLC is needed either (assuming the ATB multipliers are all even).
3. ???

whoa, that CLC actually does nothing, even or odd, by being right before the LSR.  think they typoed on order.

i experimented, and couldn't further shrink the function either.  if only there was a way to add A to X or Y. :/  i did come up with a fascinating excerpt to load the ATB Multiplier from a data table, but it was break-even at 18 bytes.

Quote
I think my proposal is fairly strictly limited to the few functions already modified to create the functionality of the patch, so it creates no unexpected incompatibility. Neither should it produce any unexpected issues for users who only care about vanilla functionality. Gutting unused parts the code does make it hypothetically less robust, but not to the point of collapse under the constraints of normal gameplay.

With that being said, this is a valid point, especially in a world where it's so easy to mod the game. This is why I'm a bit gung-ho about this as long as I don't break vanilla. My idea was to provide a .ips patch that works fine within the constraints of vanilla, for people who just want to patch and play, and supplement it with an easily modifiable source file (and ample documentation) for people who want to make mods.

yeah, your source files have been well-documented, including deletions.  you indeed had this:

Code: [Select]
;(Removed code, for reference. Removes 11 bytes)
;LSR
;LSR
;CMP #$0800
[...]

maybe add a "(Will need to be added back for a hypothetical patch that gives Morph status on turns other than the command proper.)" or similar comment.


Quote
Quote
when did free space become THAT scarce?
If not sometime in the past ten years of bug fixes, presumably sometime not too far in the future. There are many bugs left and thrift is a virtue. :P

eventually, we'll probably have to resort to ROM expansion, and patch authors coordinating (or consenting, or being too dormant to refuse ;) ) relocating many of their bugfixes together.  kind of like the PB team did.

but thrift can push this day of reckoning off for a bit.

Quote
I apologize in advance for the short novella I'm about to produce, but I have given this topic quite some thought and I feel ready share my conclusions. You may have provoked this, Assassin, but it's not directed specifically at you, although I am grateful that your rhetorical aside gave me pause to think about this.

[snip]

no prob, it was good reading.  i like seeing other patch makers' philosophies.  and hopefully, i didn't imply you were single-minded.  some of my comments could have been worded better and/or were a bit dramatic, so thanks for taking them in stride instead of getting irritated.

Quote
Free space requirements make it more difficult to maintain patch compatibility. If something can possibly fit inline, it's pretty important to make it do so, in my opinion, especially for .ips users who can't change where extra code goes. However, the more important consideration for compatibility is avoiding reaching out of scope or modifying unrelated functions, in order to avoid potential (and unexpected) code conflicts. This requirement of modular code naturally trumps free space thrift, but if it can be fulfilled, then the next order is to make the patch fit inline if possible.

yeah, my "Control attacks ignore MP cost fix" once violated your third sentence pretty badly.  i invaded a ton of nearby functions just to make my changes fit in existing areas.  i eventually optimized it years later to invade far less, but i should have been willing to use free space _until_ i could achieve that, instead of creating a monstrosity.

the thing with free space is it's fungible.  that is, N contiguous free bytes here are effectively the same as N such bytes over there (provided same bank and such).  that cuts both ways:
- on the one hand, consuming those bytes in a patch will present an eventual obstacle to ANY patch that needs free space down the road (no matter how unrelated), as they've been depleted.  that is, everybody's competing for a common resource.  in contrast, invading a neighboring function conflicts with very few of the patches out there.
- on the other hand, until free space in a bank has been exhausted, it's very easy for the later patch maker to avoid a conflict with the earlier one.  just move your code/data.  in contrast, invading a neighboring function will be a big mess in the rare case where the second author needs to change that function.

iow, a high chance of easily-avoided conflict (until exhaustion), versus a low chance of significant conflict.

Quote
With all these considerations in mind, I believe that, for now, my list of priorities looks something like this:

1) Achieving the objective of the patch
2) Patch compatibility (modular code, free space thrift, expected behaviour etc.)
3) Ease of use for end-users (patch and go)
4) Being thrifty with free space for future patches
5) Ease of use for hackers/patch authors
6) Elegance and simplicity (for personal satisfaction and bragging rights)

That last one IS a point of pride for me but it must unfortunately fall below all the others if there is a conflict. Luckily, it more often goes hand in hand with most of the other requirements.

i'd struggle too much to quantify my internal list..  but do believe that "free space thrift" has moved down it over time (though maybe with some bounces back up), as you might've guessed. :)

Quote
Finally, and most importantly, I still consider myself if not a novice, at least fresh off the boat to the world of assembly coding and patch making. Every project is a huge learning experience and so I try to push hard to work within the constraints of fulfilling all the requirements I set up for myself, or at least finding the best compromise. In most cases it would be fine to relax a little bit, if you already know what you're doing, but I'm still learning and would prefer not to afford myself the allowances that come as reward for experience just yet.

you're off to a good start.. here's to many years of it!

202
I'm sorry, what does any of this have to do with rods and spells? :offtopic:

we strayed a bit with the SMG talk, but Xenovant brought it back on-topic with the Peter North reference.  just think of Slow as being cast upon breakage...

203
Game Modification Station / Re: Patch Idea: Gau's Catharsis
« on: May 12, 2016, 02:05:35 AM »
shouldn't Gau have less Rage after a catharsis?

*folds hands* now please, take a seat on the couch.

the first post is probably the most balanced, though the other ideas are interesting too.  i'd never thought about giving him (or many other characters) a plot-based upgrade, and the dress-up event and meeting is the perfect place for it.

204
Game Modification Station / Re: WIP Patch: Permanent Morph bugfix
« on: May 11, 2016, 04:40:22 PM »
no prob.. the more testing the better, so definitely put it through the wringer.

re your prior post: yeah, the C2/0168 and C2/01B7 areas are good suspects, too.  you might be onto something with $3219.

205
maybe replying, "I preferred his earlier work." would make things less awkward?

206
Game Modification Station / Re: WIP Patch: Permanent Morph bugfix
« on: May 11, 2016, 03:38:44 PM »
odd.. my limited tests didn't have the gauge reset until after the 2nd Quick turn.  can you tell me what steps you took?

207
Game Modification Station / Re: WIP Patch: Permanent Morph bugfix
« on: May 11, 2016, 03:01:14 PM »
give this a whirl:
http://assassin17.brinkster.net/forum-posts/quick-gembox-try-fix1.asm

disclaimer: i really don't know what i'm doing.  very likely to have side effects.  EDIT: maybe the skippage should be confined to the final Quick turn.

208
Game Modification Station / Re: WIP Patch: Permanent Morph bugfix
« on: May 11, 2016, 01:49:14 PM »
magitek:

Quote
I would hope any potential hack authors interested in allowing multiple Morphers and using this patch would read the documentation I intend to write (where this caveat would naturally feature) and find out how to remedy the problem (commenting/uncommenting some lines in the included .asm file).

while i still am not a fan of the cap drop, maybe going this avenue and showing the old code would deal with any hackers who might give Morph status via alternate means.  a compromise of sorts.  i'd keep the code in and operational, but it's ultimately not my call.

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

Xenovant: have no shame; it's elusive.  i'm zeroing in a little on suspecting Function C2/091F as what'll need editing, but that's just a guess for now.

EDIT: maybe have the C2/0962 area detect whether the C2/093D area freshly set the top bit of $3AA0?  that's even more of a guess, but i think this function is going to be key somehow.

209
hey, i'm in the perfect age range.  everybody five or more years younger than me is naive and impulsive, and everybody 5+ years older is cranky and desiccated.

of course, if our parents (or nephews and nieces?) had accounts here, they'd correct us both on the one true SMG:
http://www.imdb.com/name/nm0002091/?ref_=nv_sr_2

210
looks like Seibaby (who totally doesn't post here) tackled it:
http://www.ff6hacking.com/forums/showthread.php?tid=3214

and in my day, SMG disavowed guns, only firing wooden stakes and witty barbs!