It still happens with Leadership stars and/or Str/Mag.
I think this is a problem of the newest version, because it doesn’t happen in an older version I have (2018)
It still happens with Leadership stars and/or Str/Mag.
I think this is a problem of the newest version, because it doesn’t happen in an older version I have (2018)
Must be very recent as my not-that-old version doesn’t seem to have this issue. Looking at the git, my latest version doesn’t have Gridmaster so I assume the latest PR of Sme’s branch is to blame.
Sme plz
Fixed, fix is included with pending Organization pr
There seems to be an issue with the stone spell in the skill system while the str/mag split patch is applied. I don’t know how to describe the issue, since it’s so bizarre that I think it even destroys .sav files. For the time being I made this provisional fix.
.thumb
.org 0x0
@r0,r5 - empty
@r4=attacker or defender data in battle struct
mov r0,#0x48
ldrb r0,[r4,r0]
cmp r0,#0xb5
bne notStone
stone:
add r4,#0x5A
strh r5,[r4]
mov r0,#0x14
ldsb r0,[r6,r0]
add r0,r5,r0
strh r0,[r4]
b End
notStone:
mov r0,#0x5A
strh r5,[r4,r0]
mov r0,#0x4C
mov r7,#0x14
ldr r0,[r4,r0]
mov r5,#0x2
tst r5,r0
beq IsStr
Magic:
mov r7,#0x3A
IsStr:
ldrb r7,[r6,r7]
mov r5,#0x5A
ldrh r0,[r4,r5] @current damage
add r0,r7
strh r0,[r4,r5]
b End
End:
bx r14
Replace the Str or Mag Damage.asm file with the above. What this does is check to see if the attacker is equipped with Stone (item index 0xb5). If yes, it runs the vanilla routine, otherwise it runs the new routine.
I’m not sure what a better fix for the issue would look like, but this seems to be working for now.
I believe I found the cause of this bug.
In the CanUnitWieldWeapon.s file, the ASM used in the IsUnitStatused portion of the code had a type.
I believe that Line 123, which is the operation “and r0,r1” is a typo. When compared to the original routine, this should be an “add” operation, not “and”. As such, the proper code for the routine should be as below.
IsUnitStatused:
mov r0,r4
add r0,#0x30
ldrb r1,[r0] @This is the status byte?
mov r0,#0xF @status 0xF = stone
and r0,r1
ldr r3,ItemTable
cmp r0,#3
bne PrepareExternalLoop
mov r1,#0xFF
and r1,r5
lsl r0,r1,#3
add r0,r1
lsl r0,r0,#2
add r0,r3
ldr r0,[r0,#8]
mov r1,#2
and r0,r1
cmp r0,#0
beq PrepareExternalLoop
I found another bug.
An unit that has a debuff greater than the unit’s stat makes it go over 250
This only happens with the Str/Mag Split, not in the regular version.
Jo, that guy did his training
I dare anyone to do distant counter you cowards
sorry, I reply was late.
I have confirmed that this fix works correctly.
I hope the SkillSystems github repository will be fixed.
I found a bug in ReaverSplit.
I found a bug in Rally Spectrum(Lull.s).
Thanks for reporting these! I’ve added your fixes to these three issues as well as Blademaster’s CanUnitWieldWeapon fix to a currently pending pull request.
I’ve found one problem and report it.
ifdef is required.
SkillSystem\Skills\EffectivenessSkills\EffectivenessSkills.event
Old
#include "Nullify/Nullify.event" //includes effectiveness rework installation
#include "EffectivenessRework.event" //for setting up reworked effectiveness
New
#include "Nullify/Nullify.event" //includes effectiveness rework installation
#ifdef ENABLE_SLAYER_AND_EFFECTIVENESS_REWORK
#include "EffectivenessRework.event" //for setting up reworked effectiveness
#else
#endif //ENABLE_SLAYER_AND_EFFECTIVENESS_REWORK
SLAYER_AND_EFFECTIVENESS_REWORK will be executed even if ENABLE_SLAYER_AND_EFFECTIVENESS_REWORK is disabled.
After that, the item editor.csv of Tables forcibly overwrites the contents of vanilla, so it works properly somehow.
However, SLAYER_AND_EFFECTIVENESS_REWORK consumes about 100 bytes of wasted space.
I recently updated from the October 2019 version to the February 2020 version. Some great new features! Only issue I’m running into is that every unit seems to be under the effect of “Triangle Adept”, regardless of whether they have that skill. A unit who learns it sees no benefit. This could easily be an issue on my end, but thought I’d mention it.
I think this problem is just forgetting “add r4, #0x5A”.
SkillSystem_FE8-master\Engine Hacks\Strmag\Str Mag Split\Combat
Str or Mag Damage.asm
.thumb
.org 0x0
mov r0,#0x5A
strh r5,[r4,r0]
mov r0,#0x4C
mov r7,#0x14
ldr r0,[r4,r0]
mov r5,#0x2
tst r5,r0
beq IsStr
Magic:
mov r7,#0x3A
IsStr:
ldrb r7,[r6,r7]
mov r5,#0x5A
ldrh r0,[r4,r5] @current damage
add r0,r7
strh r0,[r4,r5]
b End
End:
add r4,#0x5A @for stone <-------!!!
bx r14
This function is inserted into 0x2AB4C of FE8U.
In vanilla, “add r4, #0x5a” is issued.
The person who wrote this code forgot to increase the r4 register.
Forgetting it has affected the latter branch.
Mistakes become noticeable when branching to stone at 0x2AB62.
0802AB4C 345A add r4, #0x5a //Fixed Skill SkillSystems 20171130(スキル拡張)@0002AB4C.bin@BIN
0802AB4E 8025 strh r5, [r4, #0x0]
0802AB50 2014 mov r0, #0x14 //ResをMagicとして利用するパッチ 20171125(Install)@@EA
0802AB52 5630 ldsb r0, [r6, r0]
0802AB54 1828 add r0 ,r5, r0
0802AB56 8020 strh r0, [r4, #0x0]
0802AB58 1C30 mov r0 ,r6
0802AB5A 3048 add r0, #0x48
0802AB5C 8800 ldrh r0, [r0, #0x0]
0802AB5E F7EC FCC5 bl 0x080174ec //GetItemIndex
0802AB62 28B5 cmp r0, #0xb5 //stone <---!!!!!!!!!!!!!!!!!!
0802AB64 D101 bne #0x802ab6a
0802AB66 2000 mov r0, #0x0
0802AB68 8020 strh r0, [r4, #0x0]
0802AB6A BC08 pop {r3}
0802AB6C 4698 mov r8, r3
0802AB6E BCF0 pop {r4,r5,r6,r7}
0802AB70 BC01 pop {r0}
0802AB72 4700 bx r0
Therefore, I think that it works by just adding “add r4, #0x5A” to the end of “Str or Mag Damage.asm”.
Can you prove that bug?
or Can you send report7z?
So I was wrong, Triangle Adept is clearly adding +2 damage, so it is doing something (I thought it would also reduce the enemies Hit - kinda wish it did). But why +2? It’s as if the base Triangle bonus is being counted as +/- 30 and +/- 2 damage. This is my Weapon Triangle screen:
If I change these values, the effects in-game are doubled. So lowering to +/- 5 hit, in-game drops from +/- 30 to +/- 10.
in SkillSystems, it seems that the Weapon triangle is calculated twice.
The following settings were used for easy calculation.
(This time, I forgot to turn off the Leadership star.)
Weapon triangle: 0
Vanilla:
SkillSystems:
Weapon triangle: 10
Vanilla:
SkillSystems:
Weapon Triangle : Hit=0 Attack=0
Vanilla:
Mt 10 vs 8, Hit 56 vs 85 ,Crit 0 vs 14
SkillSystems:
Mt 10 vs 8, Hit 50 vs 95 ,Crit 0 vs 14
Weapon Triangle : Hit=10 Attack=1
Vanilla:
Mt 9 vs 9, Hit 46 vs 95 ,Crit 0 vs 14
SkillSystems:
Mt 8 vs 10, Hit 30 vs 100 ,Crit 0 vs 14
I think that is new bug.
This is because I did mistake.
When fixed the bug in ReverseWeaponTriangle, I also fixed DoubleWeaponTriangle, which is written the same way.
However, DoubleWeaponTriangle was written correctly.
It should not have been fixed.
.global DoubleWeaponTriangle
.type DoubleWeaponTriangle, %function
DoubleWeaponTriangle:
@r0=attacker
@r1=defender
push {r4-r5,r14}
mov r4,r0 @attacker
mov r5,r1 @defender
@load equipped item's weapon ability word
ldr r0,[r4,#0x4C]
ldr r1,=0x00400000
and r0,r1
cmp r0,#0
beq DoubleWTAEffectBitSet // <<--- bne DoubleWTAEffectBitSet
@check if defender has double WTA weapon too
ldr r0,[r5,#0x4C]
ldr r1,=0x00400000
and r0,r1
cmp r0,#0
bne Return @if they do, don't double WTA effect // <<--- beq Return
The FEBuilderGBA patch has been fixed.
If you update SkillSystems 20200223 after updating to the latest version, the modified version will be applied.
However, if you updated SkillSystems earlier than that, please use the hotfix patch.
Fix ReaverSplit Glitch for SkillSystems 20200223
If this patch is ready for installation, install it because it has a bug.
If it cannot be installed, the problem has already been fixed.
I wish to report a small bug. I have the most recent patch of skillsystems, and it seems equipped weapons are no longer giving the skill I told them to give. I have a silver lance trying to give the Adept skill when equipped (Skill id 0A for reference), but it doesn’t seem to work anymore. the weapon functions like normal, but it’s just not giving the skill to the unit that equipped it.