Questions about "press button" cheats

RobertLainsley1994 · 2017-03-17 04:46:52

I don't know where else to post this, and my problem's answer is not in the FAQ nor in the Knowledge Base, but if a problem like this was reported before please link me to a post that was properly solved.

My problem is with the usage of "Press Button to change parameter" cheats in the Tomodachi Collection ROM. I've checked the cheats for proper button mapping and all that, but even with every prior solution given to try to solve this problem never worked. Even with the input display turned on to make sure I was pressing the buttons, the cheats never activated. Also to note, the cheat writer was not specific where or when to press the specified button. Any suggestions?

Example:
Money 9999999 (Select)
94000130 FFFB0000
02123044 0098967F
D2000000 00000000

Last edited by RobertLainsley1994 (2017-03-17 04:48:08)

zeromus · 2017-03-17 04:58:09

use a recent build, not 0.9.11

RobertLainsley1994 · 2017-03-17 05:02:43

zeromus wrote:

use a recent build, not 0.9.11

The official download page only has 0.9.11. Can you please link me to the most recent version?

RobertLainsley1994 · 2017-03-17 05:09:50

never mind i think i got it

I'll see if it works.

RobertLainsley1994 · 2017-03-17 05:16:50

It still doesn't work, even on the most recent build. Any other suggestions?

zeromus · 2017-03-17 16:36:21

There aren't any known bugs in the cheat engine on latest build (and several were definitely fixed since 0.9.11)
There's two versions of this game. Try the other.
Cheating is hard--this is just one of many reasons why. I can't even guess where it wants you to press select. Maybe google can tell you if someone else has figured it out.

atanil · 2017-05-03 15:08:52

Are you using dynamic recompiler?
If I have it turned on these cheats are working only occasionally.

zeromus · 2017-05-03 19:10:05

I can't believe that whether a cheat uses "press button" is affected by the recompiler. Please find a cheat that's broken by the recompiler that isn't a "press button" cheat.

atanil · 2017-05-03 23:27:03

I didn't write it very clearly.
Cheats are working fine, but most of the times it doesn't register the buttons.
For example I am playing Pokemon White right now and use this code:

Overworld Codes: No Wild Pokemon (Hold R for Instant Encounter)

521A9444 D1032800
121AA440 0000E001
D0000000 00000000
521A9444 D1032800
94000130 FEFF0000
121AA440 000046C0
D2000000 00000000

I get no encounters, but with dynamic recompiler on holding R does nothing most of the time and everything works fine when it's off .

Last edited by atanil (2017-05-03 23:32:11)

zeromus · 2017-05-04 00:28:10

cheats are bugged in last official release. use a buildbot build.

fintogive · 2017-05-04 09:51:09

if i remember correctly dynamic re compiler only effects ASM codes with button activators. not pointer codes. but yes codes can be less responsive with dynamic re compiler on, last i checked. im not sure if the issues have been fixed since i dont use dynamic re compiler much.

zeromus · 2017-05-04 19:13:22

i'm skeptical. it just doesnt make sense. it's no different than running one frame at a time and tweaking memory manually.

fintogive · 2017-05-05 02:43:51

zeromus wrote:

i'm skeptical. it just doesnt make sense. it's no different than running one frame at a time and tweaking memory manually.

ok heres a perfect example:

infinite speed foward and reversed mkds (NTSC)
02072720 E3A0045F
02039988 EAFF1D88
E2000FB0 00000030
E92D0600 E59F9018
E59FA018 E589A000
E2899E0C E59FA010
E589A008 E8BD0600
E12FFF1E 01FFAE98
E350045F C3A0E45F

if you enter this code without dynamic recompler wile on a time trial mode course in the game you will be able to instantly go in reverse at infinite speeds and if you pause and restart the race. the rest of the code works as it should.

with dynamic recompler on if you enter this code wile on a time trial course nothing changes if you restart the race still nothing changes if you check the disassembler and check this address 02072720 the value is set to E3A0045F as it should be but it doesnt effect the game for some reason.

if you go back to the main title screen and go back to a course in the game the 300cc code (02072720 E3A0045F) will kick in but not the rest.
I have to reset the rom completely to get this code to work as it should in dynamic recompler

zeromus · 2017-05-05 05:47:18

OK, i see now. These cheats are program patches, and the JIT is such a big hack, it can't deal with patched programs. It makes a faint effort to insure against dysfunction by "self modifying code", yet that's what this is-- and it doesn't work.

Ideal solution is to fix the JIT. But I'm not gonna touch that.

I'm gonna fix it with a big hack: if main memory is CHANGED reset the jit entirely. This means I have to now test before having AR codes write anything to main memory, otherwise the JIT will continually rebuild (and that's very slow)

Now if a game does this on its own, the game is bugged. Well, that's jit, in a nutshell.

https://github.com/TASVideos/desmume/co … ed679e7b50

fintogive · 2017-05-31 03:36:14

well now using a pointer based code with jit causes a constant reload of jit and greatly slows the game down.

zeromus · 2017-05-31 06:34:26

Then your pointer based code CHANGES main memory every frame. Make your pointer based code not CHANGE memory every frame.

fintogive · 2017-06-01 01:55:53

thats not something i can fix with a code modifacation its how jit handles the code. its only happening with codes that try to fight the ar code as in when the code is active the game is trying to return the value the ar code has frozen back to its default state. thus jit constantly reloads.

zeromus · 2017-06-01 02:47:43

bummer. I can't fix it without learning how the jit works, which I'm not going to do. You can enter it on the github tracker if you want "rebuild jit cache when applying AR cheats only for affected memory (and also while we're at it, confirm whether self-modifying code actually works, because it doesn't seem like it does)" so we can probably never fix it.

zeromus · 2020-09-06 01:11:07

I looked into this today trying to remember what was going on, without really intending to actually fix it.

The "infinite speed foward and reversed mkds (NTSC)" example you gave --- it works right now? Did you mean there are other "pointer based code" that doesn't work? I don't have a test case right now...

fintogive · 2020-09-08 08:55:09

zeromus wrote:

I looked into this today trying to remember what was going on, without really intending to actually fix it.
The "infinite speed foward and reversed mkds (NTSC)" example you gave --- it works right now? Did you mean there are other "pointer based code" that doesn't work? I don't have a test case right now...

i was actually coming on to say thank you for putting time in to the disassembler its finally feels like a tool (I CAN SCROLL!) plus break points is extremely useful! i will check the jit for you now.

fintogive · 2020-09-08 09:10:27

wow jit is finally useful to me now! it works perfectly with the codes now awesome an age old issue finally eliminated now all thats needed is the ability to poke/copy/edit values in the ram search and i dont remember the details but just the last basic functionality for editing and creating codes.

thank you for taking the time to fix it!

AsPoke3172 · 2020-09-08 10:48:24

Huh? Cheat with JIT are works or not when pressed any button cause speed becomes slow on entire games? Hmmm.....

zeromus · 2020-09-08 19:50:19

Codes that continually rewrite the same executable memory with DIFFERENT values (possibly due to the game writing a value back) should get really slow because the executable JIT code has to be thrown away.

I can't believe this happens very often though...

https://github.com/TASVideos/desmume/issues/369
this guy says he found it. I am skeptical but he kind of has proof..

AsPoke3172 · 2022-05-26 05:22:06

After long times... Testing AR codes with JIT turned on (changed block size into 12), will results... GAME RUNNING FINE! Nooo moore sloooooowww!!!!!

zeromus · 2022-05-26 06:55:15

I have no idea why. Nothing changed the fundamental problem which was needing to rejit too much every time the program is written to by the cheat. I am suspicious and worried that something's gone wrong and the cheat processing has broken.

#1 2017-03-17 04:46:52

Questions about "press button" cheats

#2 2017-03-17 04:58:09

Re: Questions about "press button" cheats

#3 2017-03-17 05:02:43

Re: Questions about "press button" cheats

#4 2017-03-17 05:09:50

Re: Questions about "press button" cheats

#5 2017-03-17 05:16:50

Re: Questions about "press button" cheats

#6 2017-03-17 16:36:21

Re: Questions about "press button" cheats

#7 2017-05-03 15:08:52

Re: Questions about "press button" cheats

#8 2017-05-03 19:10:05

Re: Questions about "press button" cheats

#9 2017-05-03 23:27:03

Re: Questions about "press button" cheats

#10 2017-05-04 00:28:10

Re: Questions about "press button" cheats

#11 2017-05-04 09:51:09

Re: Questions about "press button" cheats

#12 2017-05-04 19:13:22

Re: Questions about "press button" cheats

#13 2017-05-05 02:43:51

Re: Questions about "press button" cheats

#14 2017-05-05 05:47:18

Re: Questions about "press button" cheats

#15 2017-05-31 03:36:14

Re: Questions about "press button" cheats

#16 2017-05-31 06:34:26

Re: Questions about "press button" cheats

#17 2017-06-01 01:55:53

Re: Questions about "press button" cheats

#18 2017-06-01 02:47:43

Re: Questions about "press button" cheats

#19 2020-09-06 01:11:07

Re: Questions about "press button" cheats

#20 2020-09-08 08:55:09

Re: Questions about "press button" cheats

#21 2020-09-08 09:10:27

Re: Questions about "press button" cheats

#22 2020-09-08 10:48:24

Re: Questions about "press button" cheats

#23 2020-09-08 19:50:19

Re: Questions about "press button" cheats

#24 2022-05-26 05:22:06

Re: Questions about "press button" cheats

#25 2022-05-26 06:55:15

Re: Questions about "press button" cheats

Board footer