Re: Parallel Worlds 1.3 Progress Thread

[Puzzledude]

this is all very strange. i absolutely entered the name as TotT, which is verified by the fact that farores isle was replaced by farores secret, granting me the rods.

In which case you should also get to the Bonus 2 rooms after Ganon.

Just use the cheat for Ocarina/Flute, and fly to number 1 and drop down to Ganon. If in second quest, you don't fall to room 0, so there is no Shrine when you go up.

[fsg]

this is all very strange. i absolutely entered the name as TotT, which is verified by the fact that farores isle was replaced by farores secret, granting me the rods.

In which case you should also get to the Bonus 2 rooms after Ganon.

Just use the cheat for Ocarina/Flute, and fly to number 1 and drop down to Ganon. If in second quest, you don't fall to room 0, so there is no Shrine when you go up.

so the hole to ganon leads to a completely different room on the underworld map in 2nd quest?

[Puzzledude]

so the hole to ganon leads to a completely different room on the underworld map in 2nd quest?

I checked. Actually you fall to room 24 in first and second quest. But in quest 1 a special ASM is telling the game to go to Shrine directly from room 24 in any case (exception). In the second quest the ASM is telling the game to go up normally to room 8 (as it is on the room-grid) and then when you reach room 0 with Trinexx, the only way to go is to the shrine (ie up from room 0), ie regular code.

I don't know why you think this is the problem. These are just 2 rooms, and bugged as well, since most emulators and probably also real hardware can not render these 2 rooms, since the first one has too much data (ie layer 1 is "masked" with layer2). Thus way too much elements in one room. The CPU choaks and makes GFX bugs, which then "scroll" to the Tronexx room. Plus Trinexx GFX can not be in the Ganon room, so when the Ganon gate opens, you have further GFX glitches. So only ZSNES emulator can render this correctly (too much elements, but not the Ganon gate). If you ask me, the second quest should be skipped in this game, since it is long enough with one quest.

[fsg]

so the hole to ganon leads to a completely different room on the underworld map in 2nd quest?

I checked. Actually you fall to room 24 in first and second quest. But in quest 1 a special ASM is telling the game to go to Shrine directly from room 24 in any case (exception). In the second quest the ASM is telling the game to go up normally to room 8 (as it is on the room-grid) and then when you reach room 0 with Trinexx, the only way to go is to the shrine (ie up from room 0), ie regular code.

I don't know why you think this is the problem. These are just 2 rooms, and bugged as well, since most emulators and probably also real hardware can not render these 2 rooms, since the first one has too much data (ie layer 1 is "masked" with layer2). Thus way too much elements in one room. The CPU choaks and makes GFX bugs, which then "scroll" to the Tronexx room. Plus Trinexx GFX can not be in the Ganon room, so when the Ganon gate opens, you have further GFX glitches. So only ZSNES emulator can render this correctly (too much elements, but not the Ganon gate). If you ask me, the second quest should be skipped in this game, since it is long enough with one quest.

i have beaten this game many many times already, and the 2nd quest offers some extra stuff i am interested in. its not weird that i wonder why this doesnt seem to work for me. i am not really bothered by glitched out gfx and stuff like that

[Puzzledude]

its not weird that i wonder why this doesnt seem to work for me.

Indeed, however it is weird that you are having problems here, since like said, the code for the second quest wasn't touched since version 1.0. The only way I could tell is to have the Rom you are using, and of course what device you are using it on.

[fsg]

its not weird that i wonder why this doesnt seem to work for me.

Indeed, however it is weird that you are having problems here, since like said, the code for the second quest wasn't touched since version 1.0. The only way I could tell is to have the Rom you are using, and of course what device you are using it on.

yes, i thought the platform might be the problem. the first time i encountered it was on snes9x 1.53, but this time it is on real console using an everdrive

[Puzzledude]

yes, i thought the platform might be the problem. the first time i encountered it was on snes9x 1.53, but this time it is on real console using an everdrive

Like said, I'm using snes9x and if on second quest, it always loads the 2 rooms after Ganon, then goes to Trinexx and then to Shrine. No idea why it doesn't work for you without looking at the Rom you are using. Real console should read the code the same way as the emulator, so ...

[fsg]

yes, i thought the platform might be the problem. the first time i encountered it was on snes9x 1.53, but this time it is on real console using an everdrive

Like said, I'm using snes9x and if on second quest, it always loads the 2 rooms after Ganon, then goes to Trinexx and then to Shrine. No idea why it doesn't work for you without looking at the Rom you are using. Real console should read the code the same way as the emulator, so ...

yes its likely to be the actual rom acting up, but may i ask which version of snes9x you tested this on? and its not guaranteed it will work the same on console (unless its tested on console of course)

[Puzzledude]

yes its likely to be the actual rom acting up, but may i ask which version of snes9x you tested this on?

1.51, but now also on 1.55 also goes to bonus rooms. Zsnes also goes to 2 bonus rooms.

and its not guaranteed it will work the same on console (unless its tested on console of course)

Yes, however this should work.


In fact, if you use the walk through wall cheat it doesn't want to go to Shrine at all. I always goes to that additional dungeon room.

I now also tested on snes9x 153 and snes9x 153 8MB by Fusoya. Same result. If name is TotT and gate open, you go to additional rooms, if name is anything else, you go to Shrine, but only if the Ganon gate opens. If you are in quest 1 and use the walk through walls cheat, it will always go to additional room instead of Shrine, since the code is triggered by the Ganon gate (opened or not).

So irronically it is a lot easier to come to the additional room than to the Shrine.

[Puzzledude]

Try this: use any version of snes9x for PC. Open the Rom in hex: remove Ganon by going to 4D91D: and change D6 to 42= change Ganon to (glitched) green soldier.
Load Rom, enter name TotT. Cheat for Flute+bird= 7EF34C03 and for hammer:7EF34B01.

Start the game and fly to 1, drop down, use hammer to defeat the soldier: Ganon gate will now open. Go in and see what happens: you should be in bonus room.

Repeat the same if name is not TotT, you should be in Shrine.

This is how I have it. If this works, than you know that the console compatibility is the problem.

[fsg]

so i think i know what might be causing it now. when you go to the fat fairy in the 2nd quest, there is this hylian shield (which you do not have) that you can throw into the pond, which turns into a blue mail that the fat fairy gives back to you, so essentially a free blue mail. obviously i exploited this.
this glitch is 2nd quest exclusive, and my guess is that whatever code is supposed to be executed (or not executed) when you transition north from ganon does not work if you do this blue mail glitch.
i tried EGing to ganon and transition north to test this, but this proved to be much more problematic than what i expected (you fall into the holes on the ganon screen when you try this).
i could of course try this on emulator, but i wanted to make sure this have nothing to do with the platform im on

[Conn]

Before this turns into something disgusting, stop that now - the both of you!
I really do not want to have such things here on the forums. Really not.

So focus on the main issue, if reported bugs occurs only on real hardware (as it appears to be), there's nothing the devs of  PW can do. sd2snes bugs are impossible to trace (there is no debugger), it would be try and error.
fsg, did you try on emulator (maybe with the same srm) if the bugs occur there as well then?

[fsg]

Before this turns into something disgusting, stop that now - the both of you!
I really do not want to have such things here on the forums. Really not.

So focus on the main issue, if reported bugs occurs only on real hardware (as it appears to be), there's nothing the devs of  PW can do. sd2snes bugs are impossible to trace (there is no debugger), it would be try and error.
fsg, did you try on emulator (maybe with the same srm) if the bugs occur there as well then?

it is unlikely to be related to the hardware, as i had the exact same thing occur on snes9x (i didnt own a snes the first time this happened)

[fsg]

Maybe we can open up a court case here:
you now have video proof of this being glitched, while in my case no glitches.

this is very odd indeed. but may i ask what this shield even is? it is just some odd item which i do not possess that i can throw into the pond. or did you actually use a cheat to obtain the shield before throwing it into the pond?

[Conn]

can you maybe provide the srm in which the bugs occur? In this case people could trace which ram address leads to that bug.

[Puzzledude]

Before this turns into something disgusting, stop that now - the both of you!
I really do not want to have such things here on the forums. Really not.

So focus on the main issue, if reported bugs occurs only on real hardware (as it appears to be), there's nothing the devs of  PW can do. sd2snes bugs are impossible to trace (there is no debugger), it would be try and error.
fsg, did you try on emulator (maybe with the same srm) if the bugs occur there as well then?

Yes, if you ask me some posts could be deleted in this topic.

[fsg]

Maybe we can open up a court case here:
you now have video proof of this being glitched, while in my case no glitches.

so by looking closely at your pictures, it seems like you actually have the hylian shield in your possession before throwing it into the pond. the glitch here is that i DONT have this shield at this point (you can see from the video that i have the kokiri shield), and this is the reason why i decided to throw it into the pond the very first time i went there in 2nd quest. try going to the pond in 2nd quest without having the hylian shield and see what happens

can you maybe provide the srm in which the bugs occur? In this case people could trace which ram address leads to that bug.

what exactly is the srm, and how do i provide it?

[Puzzledude]

Ok, everyone, sorry about the drama.

The fsg has won this case.
Indeed there's a glitch there, but not if you throw in the Shield (regular ones anyway).

The glitch is this (this however was not explained properly):
have Kokiri shield and have tunic-1, then
in the big fairy's pond, there is indeed an option to throw in Kokiri shield, which is fine, but after Moopearl a fake Shield-2 appears in the list of possible items to throw in (so 2 shields can be thrown in). And if you throw in this, you indeed receive the tunic-2, and for some unknown reason you than indeed show up in the Shrine after Ganon.

So I admit my defeat here.
Unfortunately nothing I can do, to fix it obviously, since this is ASM related.

Also, this is not hardware/device/sd2snes based. The bug is on all devices (emus, flashcart, real cart etc).

[fsg]

im glad it has been resolved.
the funny thing is that i discovered this blue mail glitch the first time i played the 2nd quest, but didnt report it because of selfish reasons (lol). i kinda wanted this easy access blue mail to be there.
how ironic that it was the cause of all the problems

[Puzzledude]

im assuming this image comes from the chat of someone playing one of your hacks, and i am not one of these people.

Yes, it's from the Cube hack stream, but I have to say this was to be expected from the stream chats and you tube comments.

what exactly is the srm, and how do i provide it?

SRM is short for SRAM which is short for Save RAM or the regular SNES save. And you can not obtain it from a console, since you don't have access to it (unless sd2snes). But no need since the bug is on emus too, so we just use a savestate .000 file to load the exact moment of the game before tracing in Geiger's debugger.

Otherwise you would need it to be able to fix sd2snes exclusive bugs.

[Conn]

Yes, if you ask me some posts could be deleted in this topic.

I deleted some of the posts, and am glad no further drama emerged here

The fsg has won this case.
So I admit my defeat here.

It's not about win and loose and defeated... it was a weird bug and a misunderstanding since it couldn't be reproduced easily. Believe me, I had the weirdest bug reports already and none was trolled.

Anyways, though I'm surely able to trace down that bug now with the receipt how to reproduce, I know PW too little - so I hope qwertymodo will look into that! If there are problems fixing this bug, come back at me
But if you like, you could provide me a Geiger savestate so I can check a bit why the fake shield appears.

[Conn]

I do know now what's going on:
There is an unused item ram slot (7ef358) - Euclid used this as a flag that you're in second quest (TotT), 7ef358-08. Therefore this bug only occurs in second quest.

You can throw this "flag" as an item (2nd shield) into the fairy pond, (no idea why the code interprets it as blue mail then), but the fairy  will zero 7ef358 when returning it, so you will be again in first quest. That's why it affects the other stuff reported..

I will try to figure out an easy asm fix...

[Conn]

Ok, here would be my fix suggestion:

hook1 (incrementing item):
$8E/CDFD BF 40 F3 7E LDA $7EF340,x[$7E:F358] A:CD01 X:0018

hook2 (decrementing item):
$8E/CDD6 BF 40 F3 7E LDA $7EF340,x[$7E:F358] A:CD02 X:0018

freespace:
CPX #$18  ; are we at 7ef358?
BNE $03    
LDA #$00  ; if yes, set Accumulator value to 00 to skip
RTL
LDA $7EF340,x ; if X is not 18, execute native code
RTL

So attached code will do:

Code:

07/4DFD (snes 8E/CDFD):
22 20 81 A0 ; JSL $a08120 (some freespace)

07/4DD6 (snes 8E/CDD6):
22 20 81 A0 ; JSL $a08120  (some freespace)

10/0120 (snes a0/8120):
e0 18; cpx #$18
D0 03 ; BNE $03
A9 00 ; LDA #$00
6B;      RTL
bf 40 f3 7e ; lda $7ef340,x
6B; RTL

Attachments

pw_fairyfix.zip You don't have permission to download attachments.(1 Kb) Downloaded 8 times

[Puzzledude]

I do know now what's going on:
There is an unused item ram slot (7ef358) - Euclid used this as a flag that you're in second quest (TotT), 7ef358-08. Therefore this bug only occurs in second quest.

You can throw this "flag" as an item (2nd shield) into the fairy pond, (no idea why the code interprets it as blue mail then), but the fairy  will zero 7ef358 when returning it, so you will be again in first quest. That's why it affects the other stuff reported.

Thank you for figuring this out. It makes sense now. So you litterally "throw away" the second quest and go to first quest again.

I should have seen this because of this:
in second quest a big script "2nd quest" is in the menu in the right-upper corner


But after you throw the false shield away, the script is gone:
thus you are back in the first quest, this is why you exit in shrine and this is why the ending credits will play out as if in first quest (everything else will be 1st quest also, thus you can return to light world with mirror and also enter the Farore's dungeon again and get the Hookshot):

[Puzzledude]

07/FDF4 (snes 8E/CDFD):
JSL $a08120 (some freespace)

07/4DD6 (snes 8E/CDD6):
JSL $a08120  (some freespace)

10/0120 (snes a0/8120):
CPX #$18
BNE $03
LDA #$00
RTL
LDA $7ef340,x
RTL

Great, this fixes it.
Looks like this in hex.
PW
Big Fairy fix:

74DD6,
BF 40 F3 7E --> 22 20 81 A0

74DFD,
BF 40 F3 7E --> 22 20 81 A0

100120, write new code
E0 18 D0 03 A9 00 6B BF 40 F3 7E 6B