øAslickproductions.org/forum/index.php?PHPSESSID=5f0fck550j2m4m2fpbtkj2vkm1&action=profile;u=25;area=showposts;start=360e:/My Web Sites/Slick Productions - FFIV Message Board/slickproductions.org/forum/indexccd1-2.htmlslickproductions.org/forum/index.php?PHPSESSID=5f0fck550j2m4m2fpbtkj2vkm1&action=profile;area=showposts;u=25e:/My Web Sites/Slick Productions - FFIV Message Board/slickproductions.org/forum/indexccd1-2.html.zxH›h^ÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÈ ,/ËþOKtext/htmlISO-8859-1gzip0|ÖËþÿÿÿÿÿÿÿÿWed, 11 Mar 2020 08:19:44 GMT0ó°° ®0®P®€§²ð®G›h^ÿÿÿÿÿÿÿÿU&Ëþ 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

361
Quote
A Nut Eater attacks a character with defense 0. Regular attacks do 10 damage. While using defend it's 5 damage. While using defend in the back row it's still 5 damage (it should be 2).

"should be" by your standards.  i don't intend that as snippy; just mean there are reasons to think otherwise.

Quote
So yeah, minimum M still seems to be 1. It would be really nifty if it could go lower, like 0.125 (1/8) in the more extreme cases (armor buff, defend, back row).

this goes to the heart of *why* Square had Minimum M in the first place.  it's fairly obvious they were trying to avoid low M causing 0 damage.  instead, there is a floor that's usually well above that (depending on Attack and Defense values).  the question is: how averse would Square have been to values that aren't zero, but are ridiculously close to it?  if having nonzero damage were their only goal, they could simply have given the damage value itself a floor of 1 -- but they chose to modify M instead.

besides, pretty much every M formula already has a constant value at the end (e.g. " + 2"), so starting M is _already_ nonzero.  the fact that Square waited until late in the damage calculation to ensure its minimum is 1 strongly suggests that they wanted to limit the effects that the divide-by-2 modifiers would have on low-end damage.

from what i've read of the Fractional M patch, its purpose was: as an attacker progresses from the low end of their damage range to the higher, to smooth the ascent of this damage, instead of it moving up in "steps".  doing away with the lower bound in addition to this is a bit of a reinterpretation of the patch.  i don't recall seeing anything in the patch Readme or the old thread that says the goal was to change the damage range -- just to modify the progression through it.

i'm not saying your interpretation is wrong.  simply that there's enough evidence to the contrary of it, and that if Square's only concern were avoiding zero damage, they would have taken a more direct and less aggressive approach.  saying an instance of Nut Eater damage "should be" lower is subjective, at the least.

362
Quote
Well, Carcursers have 75 HP. If you use a Bolt spell it will do 70 damage instead of 75 because of Atk/2*2. If it did 75 damage it would always kill the Carcursers in one hit (which I think might have been the developers' intention).

how would they know what level the caster of Bolt was going to be?  this theory seems unlikely; if the developers wanted a specific reaction to a spell or element, they would've done something like "React:Elemental Attack:Element:" in the monster script.

363
Quote from: Jorgur
One thing I noticed just now is that fractional M doesn't seem to apply to monsters' Fight command.  This means that if you are using Defend/back row and the monster's M is 1, you will still take full damage.

that's probably Minimum M kicking in.  the original game would set it to 1 if it was 0.  accordingly, our patch gives it a floor of 128 to behave equivalently, as a division by 128 occurs later.

the fractionalizing still works as expected for higher M values, because /2 modifiers won't bring them below the minimum.

364
ahh..  that's what i get for not going through the simple steps of 15 DIV 2 * 2 , and just assuming the modifiers cancelled. :/

Quote
Attack  = 15  + (0..2) = 15, 16 or 17

like Jorgur said, i don't think it'll ever be 17 here.  15 DIV 8 = 1.

365
do you ever get 76 in there if you cast a bunch of times, or is it only the two values?  out of curiosity, what is the damage range in the original game?  i realize it's not supposed to match; i'm more interested in the difference between high and low (again, amongst a bunch of castings).

366
Gaming Discussion / Re: FF6 - Magicite item RNG
« on: December 24, 2014, 10:59:48 PM »
i doubt it.  it uses the Bank C2 RNG, whose table index isn't incremented only by RNG calls, but also by C2/0019 on very regular increments (if not every frame, then still many times a second).

then again, the battle RNG has still been known to fall into "ruts".  see the old Mnrogar's posts where they were fighting Storm Drgn on a certain Battle Speed.

367
if the main Fractional M patch and the standalone Damage Overflow patch both hook C2/9419, but they have it call two different places, how can they be compatible with each other?

368
making the Damage Overflow fix available separately was a good idea, as was including source code with these patches.

Quote
I am sharing here then a patch for the Damage Overflow compatible with every version (J and RPGe, with or without Fractional M) and a delta for the Fractional M patch which doesn't include the Damage Overflow, but is compatible with it.

however, how are you able to have the standalone Damage Overflow patch compatible with the Fractional M patch?  they both edit C2/9419, right?  i suspect that if somebody wants both of these changes in effect, they'll need a special edition of Fractional M that has Damage Overflow built-in.

369
Quote
This delta adds "Minimum M set to 128" and "Fixed Overflow". Still wondering if assassin found a scenario where 0 < M < 128.

i think it's possible.  if the attacker has very low Level and low relevant stat(s), you get an M of something like: double-digit-number + 256.  or hell, for Fists, it simply is 256.  should you perform two or more M = M / 2 modifications in later functions, you can get a sub-128 M.

please doublecheck, though.  i might be missing something.

370
Code: [Select]
C2/940C: A5 52        LDA $52
C2/940E: D0 01        BNE $9411
C2/9410: 1A           INC         (Minimum M = 1)
C2/9411: 20 B2 01     JSR $01B2   (Minimum M = 128) <--- First change
C2/9414: 85 2C        STA $2C

just replace "INC / JSR $01B2" with "LDA #$0080", and make the BNE skip 3 bytes.  you don't want to be increasing M here unless it was below the floor.

i'm debating whether the M check should be for == 0, or for < 128.  something about the latter seems like a safer catch-all, but it shouldn't be necessary unless earlier code was botched or missed.

Quote
I am assuming accumulator is set to 8 bits in C2/8A11, am I right?

yep.

371
Code: [Select]
------------------------------------------
32 bit ($2E > 9999) comparator
------------------------------------------
D0/xxxx: A6 2F        LDX $2F
D0/xxxx: D0 07        BNE $xxxx
D0/xxxx: A6 2E        LDX $2E
D0/xxxx: E0 0F 27     CPX #$270F
D0/xxxx: 90 03        BCC $xxxx   (jump to no)
D0/xxxx: A2 0F 27     LDX #$270F  (Max Damage = 9999)
D0/xxxx: 6B           RTL

that first line should be: LDX $30

372
Code: [Select]
C2/849B: A5 26        LDA $26
C2/849D: 20 BB 01     JSR $01BB    (Level * Magic Power)/128
C2/84A0: 18           CLC
C2/84A1: 65 52        ADC $52
C2/84A3: 85 52        STA $52      (M = M + (Level * Magic Power)/128)

the Fractional M patch doesn't touch this code.  this means that Rune weapons will have essentially no boost to M.  (but test in case i'm wrong.)

solution: NOP out C2/849D.

373
Thieves would get a damage boost, but I would argue that they don't need it.

Not really. The STR and AGL / 256 makes knifes based weapons worst also to Thieves (I explain it in Reply #54)

Quote from: reply #54
Well... it seems nobody likes my approach of using the average of Strength and Agility.

That's a good thing as the Agility is only higher than Strength in few classes and these classes are in fact the ones which use the knives... performing an average is harmful to these classes as it is to the fighter classes so I better dismiss this approach.

an average might be lower than pure Agility would be, but it still boosts these classes' damage relative to what's in the original game.

Quote
- Assassin's
    M = (STR * Lv)/256 + (AGL * Lv)/256 + 2
   A little bit unbalanced formula. Underpowered and with huge jumps in damage increments.

well, doing ((STR + AGL) * Lv) / 256 could smooth things out.  same deal with any other method in the list that suffers from damage jumps.

374
another thing regarding Fractional M:

Code: [Select]
C2/8A0D: A6 52        LDX $52
C2/8A0F: D0 15        BNE $8A26
C2/8A11: E6 52        INC $52      (Minimum M = 1)

Code: [Select]
C2/940C: A5 52        LDA $52
C2/940E: D0 01        BNE $9411
C2/9410: 1A           INC         Minimum M = 1
C2/9411: 85 2C        STA $2C

because the patch uses 128M, we should probably be giving it a floor of 128 here in order to behave equivalently to the original game.

i think the first check is there because the second check never saves the updated M in $52, and Square might want to use that updated M value for Healing.  however, that raises the question why C2/881D doesn't enforce Minimum M in a similar manner.  if "Check Target Can't Evade vs. Attack" passes there, M*2 winds up with a floor of 0, brought up to 1 by calling C2/93F8.  in contrast, if that check passes in C2/8A05, M*2 winds up with a floor of 2.  subtle difference, but odd.

375
but it's unlikely Square had Agility in that function by accident.  so imho, the best fix tries to figure out in what capacity Agility would've likely been used, while still retaining balance.
It's not quite that simple, you cannot retain balance if you include Agility. By default, in vanilla (unpatched) FF5, Knife based weapons do the exact same damage as Sword based weapons if they have the same attack power (because of the bug, knives sometimes do a little extra). It's a very simple, understandable way to calculate damage - if the weapons have the same attack power, they do the same damage. If, on the other hand, you patch the ROM to include Agility, it becomes a lot more complicated.

by "retaining balance", i meant have similar capabilities between the weapon types, i.e. similar damage ranges.

Quote
Take as an example a Knight wielding a Mythril Sword. You happen to find a MageMasher knife. Normally the two weapons do the exact same damage, but if half the Strength influence is replaced by Agility, the MageMasher will end up doing less damage (because Knight has low Agility).

so this weapon would favor Thieves..  don't high-Strength jobs already have weapons that favor them (or are outright exclusive to them)?

Quote
Quote
without an extra term in the formula, Knives are more of a poor man's Swords.  they don't really have anything to set them apart.
This is not true. According to instructrtrepe's algorithm guide, all knife based weapons halve the enemy evade stat when used. This alone gives knives an advantage over swords.

oops, i forgot about that.

Quote
Quote
I'm almost persuaded by the argument, but what do the damage numbers look like at each level with either approach?
With (Level*Strength)/128 + 2 (or vanilla FF5), knives do the exact same damage as swords of the same attack power. Every 3-6 levels the attacks become a bit stronger.

well, not the exact same.  Knives get "(0..3)" added to Attack, while Swords get "(0..(Weapon Attack)/8))" added.  the latter is generally higher.