Posts for finalfighter

1 2
8 9
Posted: 8/21/2015 4:11 PM
Quote
finalfighter
Experienced Forum User, Published Author, Active player (276)
Joined: 4/20/2004
Posts: 219
Location: Japan,Tokyo
NES TAS of 2020 Glitchy TAS of 2020 TAS of 2010 NES TAS of 2010 Glitchy TAS of 2010 NES TAS of 2009
If it isn't an error of production of real NES, that's interesting. I have not seen the glitch condition with which the "stage number"=$2A changes yet. (The stage of a recovery is changing, so it isn't glitch of the "map number".=$14,$15,$440-45F)
Posted: 8/13/2014 11:15 AM
Quote
finalfighter
Experienced Forum User, Published Author, Active player (276)
Joined: 4/20/2004
Posts: 219
Location: Japan,Tokyo
NES TAS of 2020 Glitchy TAS of 2020 TAS of 2010 NES TAS of 2010 Glitchy TAS of 2010 NES TAS of 2009
I have not discovered such a glitch yet. However, it may be interesting when there is a glitch which a game crashes or resets.
Posted: 6/17/2014 2:23 PM
Quote
finalfighter
Experienced Forum User, Published Author, Active player (276)
Joined: 4/20/2004
Posts: 219
Location: Japan,Tokyo
NES TAS of 2020 Glitchy TAS of 2020 TAS of 2010 NES TAS of 2010 Glitchy TAS of 2010 NES TAS of 2009
Although I have also already tested the part using LuaScript, mad of ProcessCounter do not appear at Wily2 Boss. http://www.yuko2ch.net/rockman/rm2delaytest.lua Because, next screen exists. Then, we cannot use ACE there. Although I tested the part which reset caused by DelayScroll(Example.(QuickManStage(=Wily5Map),Wily3(BossMap),Wily4(BossMap))), ACE was possible only in Wily3.
Posted: 6/6/2014 6:16 PM
Quote
finalfighter
Experienced Forum User, Published Author, Active player (276)
Joined: 4/20/2004
Posts: 219
Location: Japan,Tokyo
NES TAS of 2020 Glitchy TAS of 2020 TAS of 2010 NES TAS of 2010 Glitchy TAS of 2010 NES TAS of 2009
Mechanism of DoubleObjectFFGlitch http://www.yuko2ch.net/rockman/doubleobjectffglitch.txt about DoubleObjectFFGlitch(in Japanese) http://ch.nicovideo.jp/TASVideos/blomaga/ar535617 in IceManStage ObjectNumber=$23(counter) inGutsManStage ObjectNumber=$17(2nd controller) If ObjectNumber=4Aor75orF5 , StageClear is happend.
Posted: 5/24/2014 7:49 PM
Quote
finalfighter
Experienced Forum User, Published Author, Active player (276)
Joined: 4/20/2004
Posts: 219
Location: Japan,Tokyo
NES TAS of 2020 Glitchy TAS of 2020 TAS of 2010 NES TAS of 2010 Glitchy TAS of 2010 NES TAS of 2009
Good news, I success DelayMagnetBeamAdapter(Object42) without ACE. :P https://www.youtube.com/watch?v=ntbgATKvhp8 Then , I have some idea of all stage clear TAS. ->I(Object42->get MagnetBeam->Object75->DelayStageClear) ->F(using MangetBeam TAS) ->B(ObjectF5->DelayStageClear) ->G(using MangetBeam TAS) ->C(using MangetBeam TAS) ->E(using MangetBeam TAS) ->W1(Object75->DelayStageClear) ->W2(Object5D->DelayStageClear) ->W3(using MangetBeam TAS) ->W4(using MangetBeam TAS) ->LastBoss Magnet beam can be used by ElecManStage. :) Although we would like to make a fastest TAS, ("game end glitch") I would like to also make an entertainment TAS!(any% or "glitched") http://tasvideos.org/2921S.html This course may update this TAS? Do you think about this course ? ;)
Posted: 5/19/2014 11:11 AM
(Edited: 5/19/2014 3:42 PM)
Quote
finalfighter
Experienced Forum User, Published Author, Active player (276)
Joined: 4/20/2004
Posts: 219
Location: Japan,Tokyo
NES TAS of 2020 Glitchy TAS of 2020 TAS of 2010 NES TAS of 2010 Glitchy TAS of 2010 NES TAS of 2009
To be sure, it is technical. The call of the worng ROM address which a glitch causes exists also in many other games of any% TAS. (The call of a debug mode is an intelligible example. ) And the case where worng RAM address is called also exists in other games of any% TAS. Some case, reset may be avoided fortunately. Then, I think definition of ACE, only when the code is intentionally written to the memory(RAM). Though NES only... I agree with the opinion of feos. :) 3 branches, I think , - ACE (this submission) call RAM address,write codes to RAM , fastest Ending. - glitched (current run) call ROM address only, final Boss killed. - 100% run (all play, Deign's run) DelayStageClear is not included , play full of all stages
Posted: 5/19/2014 7:45 AM
Quote
finalfighter
Experienced Forum User, Published Author, Active player (276)
Joined: 4/20/2004
Posts: 219
Location: Japan,Tokyo
NES TAS of 2020 Glitchy TAS of 2020 TAS of 2010 NES TAS of 2010 Glitchy TAS of 2010 NES TAS of 2009
I think that the kind of object and used glitch is important. *kind of glitch DelayObjectGlitch(Bank5) was used by my old TAS. DelayObjectFFGlitch(Bank2) was used by my new TAS. DelayObjectFFGlitch(Bank2) differs from DelayObjectGlitch(Bank5). And the places which interrupt NMI also differ. If I am in the place where ObjectFF doesn't appears, DelayObjectFFGlitch cannot be executed. And if I cannot use DelayObjectFFGlitch, Object55 doesn't appear in BombManStage and IceManStage. *kind of object ObjectF5(DelayStageClear) using DelayObjectGlitch used by old TAS execute $C0BD. Object55 using DelayObjectFFGlitch used by new TAS execute $0600. Though it is difficult to determine the definition of ACE. :( Perhaps, I think that a ROM address and a RAM address using glitch are important. any%: worng ROM address($8000->$FFFF) executed by glitch of TAS (like old TAShttp://tasvideos.org/2921S.html) ACE : worng RAM address($0000->$7FFF) executed and the code of the RAM was executed by glitch of TAS (like new TAShttp://tasvideos.org/4308S.html) I think that anyone would like to make any% which acquired MagnetBeam in BombManStage(using DelayObjectGlitch) in the future. Probably it will show entertainment. What do you think about it? :)
Posted: 5/14/2014 1:26 PM
Quote
finalfighter
Experienced Forum User, Published Author, Active player (276)
Joined: 4/20/2004
Posts: 219
Location: Japan,Tokyo
NES TAS of 2020 Glitchy TAS of 2020 TAS of 2010 NES TAS of 2010 Glitchy TAS of 2010 NES TAS of 2009
https://www.youtube.com/watch?v=JCGoJB3g6Xk Visualize Memory
Posted: 5/13/2014 8:00 PM
Quote
finalfighter
Experienced Forum User, Published Author, Active player (276)
Joined: 4/20/2004
Posts: 219
Location: Japan,Tokyo
NES TAS of 2020 Glitchy TAS of 2020 TAS of 2010 NES TAS of 2010 Glitchy TAS of 2010 NES TAS of 2009
Thanks Mr.feos. I understand very much. :)
Posted: 5/13/2014 7:25 PM
Quote
finalfighter
Experienced Forum User, Published Author, Active player (276)
Joined: 4/20/2004
Posts: 219
Location: Japan,Tokyo
NES TAS of 2020 Glitchy TAS of 2020 TAS of 2010 NES TAS of 2010 Glitchy TAS of 2010 NES TAS of 2009
Old TAS 1:ObjectF5 calls worng address($C0BD). 2:DelayStageClear NewTAS: 1:Object55 call worng address($600). 2:The code written by us is executed.($600) Perhaps, I think that these have a difference of genre. :)
Posted: 5/13/2014 6:29 PM
Quote
finalfighter
Experienced Forum User, Published Author, Active player (276)
Joined: 4/20/2004
Posts: 219
Location: Japan,Tokyo
NES TAS of 2020 Glitchy TAS of 2020 TAS of 2010 NES TAS of 2010 Glitchy TAS of 2010 NES TAS of 2009
>Mr.feos I am confused to the definition of Ending. I consider 5 patterns to your guidelines. :) 1:All Stage Clear->Wily defeated->goto Ending 2:Ice->DelayGlitch->jump to WilyStage->Wily defeated->goto Ending 3:Ice->DelayGlitch->$31(stage) is 0x0B(Ending)->Death->goto Ending 4:Ice->DelayGlitch->$5D is mad(We get all arms)->goto Ending 5:Ice->DelayGlitch->goto Ending(this case)
Posted: 5/13/2014 4:15 PM
Quote
finalfighter
Experienced Forum User, Published Author, Active player (276)
Joined: 4/20/2004
Posts: 219
Location: Japan,Tokyo
NES TAS of 2020 Glitchy TAS of 2020 TAS of 2010 NES TAS of 2010 Glitchy TAS of 2010 NES TAS of 2009
We can set whatever value we like to $5D ($5D = arm flag, or boss defeat flag). (However, we have to sacrifice many frames to rewrite 4 more bytes of RAM into some code.) But I don't think that it's necessary for the game completion, and I think that the condition for the game to be completed is watching the ending. The reason why I think so is that any other TAS'es didn't set (or even care about) flags for the second play until now. For example, Super Mario Land 2 : The 6 coins flag are not set. :)
Posted: 5/8/2014 6:11 PM
Quote
finalfighter
Experienced Forum User, Published Author, Active player (276)
Joined: 4/20/2004
Posts: 219
Location: Japan,Tokyo
NES TAS of 2020 Glitchy TAS of 2020 TAS of 2010 NES TAS of 2010 Glitchy TAS of 2010 NES TAS of 2009
I and the Mr.Pirohiko found the ACE code. We can perform DelayEnding at GutsTank fight. :)
Posted: 5/6/2014 8:06 PM
Quote
finalfighter
Experienced Forum User, Published Author, Active player (276)
Joined: 4/20/2004
Posts: 219
Location: Japan,Tokyo
NES TAS of 2020 Glitchy TAS of 2020 TAS of 2010 NES TAS of 2010 Glitchy TAS of 2010 NES TAS of 2009
In GutManStage , I found the case of glitch where $009D was called. (not rare) http://www.yuko2ch.net/rockman/rm2_9D_called.fm2 Still, it is not delay ending. Some codes can be executed , however, it is difficult to call address directly. Because , $9C-$A6 = energy of arms. = #$00-#$1C If we can call $C269, it will be able to go to ending. However, there is still no good idea for ope-code. I am waiting for the good idea. :) The conditions of good ending. ($2A = $0D , JMP $C215) or (JMP $C269) And , I am investigating whether the call of a different address also occurs.
Posted: 5/5/2014 3:57 PM
Quote
finalfighter
Experienced Forum User, Published Author, Active player (276)
Joined: 4/20/2004
Posts: 219
Location: Japan,Tokyo
NES TAS of 2020 Glitchy TAS of 2020 TAS of 2010 NES TAS of 2010 Glitchy TAS of 2010 NES TAS of 2009
When I fight with GutsTank, if I cause DelayRightScroll , reset is caused. It is based on the call of the wrong address. I have tried by this LuaScript. This LuaScript is BOT which causes DelayRightScroll. http://www.yuko2ch.net/rockman/rm2reset.lua Some cases. 1:reset 2:fleeze 3:The beginning of a stage and reset http://www.yuko2ch.net/rockman/rm2reset6.fm2 http://www.yuko2ch.net/rockman/mm2reset.mp4 I am investigating the bad condition. :)
Posted: 5/1/2014 7:24 PM
Quote
finalfighter
Experienced Forum User, Published Author, Active player (276)
Joined: 4/20/2004
Posts: 219
Location: Japan,Tokyo
NES TAS of 2020 Glitchy TAS of 2020 TAS of 2010 NES TAS of 2010 Glitchy TAS of 2010 NES TAS of 2009
Good news, we confirmed that DelayEnding was possible. I already created three Lua files for assistance :) Then , I and Mr.Pirohiko began to make new TAS of Rockman1 !
Posted: 4/30/2014 10:05 PM
Quote
finalfighter
Experienced Forum User, Published Author, Active player (276)
Joined: 4/20/2004
Posts: 219
Location: Japan,Tokyo
NES TAS of 2020 Glitchy TAS of 2020 TAS of 2010 NES TAS of 2010 Glitchy TAS of 2010 NES TAS of 2009
I and the Pirohiko are finding a breakthrough of DelayObject55 or D5. If our investigation is right, we will start new TAS. :)
Posted: 4/29/2014 8:12 PM
Quote
finalfighter
Experienced Forum User, Published Author, Active player (276)
Joined: 4/20/2004
Posts: 219
Location: Japan,Tokyo
NES TAS of 2020 Glitchy TAS of 2020 TAS of 2010 NES TAS of 2010 Glitchy TAS of 2010 NES TAS of 2009
A promising object 
Delay
Object  CallAddress
50,D0  0390 
53,D3  00AD 
55,D5  0600 = ObjectPosY
5A,DA  00A0 
60,E0  0660 = ObjectYSpeedFraction
64,E4  0390 
68,E8  0640 = ObjectFireDelay
69,E9  00A9 
6C,EC  04C0 = ObjectXSpeed
74,F4  04E0 = ObjectXSpeedFraction
78,F8  04C0 = ObjectXSpeed
7F       04E0 = ObjectXSpeedFraction
It is interesting if these are "0x20"(JSR) or "0x4C"(JMP)- Example: If DelayObject55 is appear , $600 $601 $602... executed. ForExample, If $602(RockBuster1Y) $603(RockBuster2Y) $604(RockBuster3Y) = 4C XX YY then JMP $YYXX. If $YYXX is the address of game clear , we may get ending. We need to investigate whether a suitable value can be adjusted. We can get object DelayObject00-FE easily on an IceManStage. (BombManStage is difficult :P)
Posted: 4/27/2014 3:05 PM
Quote
finalfighter
Experienced Forum User, Published Author, Active player (276)
Joined: 4/20/2004
Posts: 219
Location: Japan,Tokyo
NES TAS of 2020 Glitchy TAS of 2020 TAS of 2010 NES TAS of 2010 Glitchy TAS of 2010 NES TAS of 2009
I need to investigate the DelayObject of reset. Because, the data of a RAM address may be executed as processing like SMB3 by DelayObject. I have not investigated yet. However, I looked at execution of instructions(6502) from the data of RAM address by some DelayObjects with the past. This is an interesting thing. It will become interesting if JSR [$XXXX] can be executed. JumpAdress list of DelayObject. http://www.yuko2ch.net/rockman/JumpAddressList.txt I need to investigate DelayObject4B-DelayObject7F. In other words , I will try to investigate the possibility of "DelayEnding" in BombManStage or IceManStage.
Posted: 2/18/2014 4:59 AM
Quote
finalfighter
Experienced Forum User, Published Author, Active player (276)
Joined: 4/20/2004
Posts: 219
Location: Japan,Tokyo
NES TAS of 2020 Glitchy TAS of 2020 TAS of 2010 NES TAS of 2010 Glitchy TAS of 2010 NES TAS of 2009
Since Shinryuu is very busy, Our TAS is still BubbleMan stage. However, some saving has already been performed. Mr.aglasscage, please cooperate in our TAS if you hope. :) Intersting. >hi.fm2 When you are in WilyStage1, you cannot go to the map 255. Because the wall is on a left end on the WilyStage 1. >testagain.fm2 Instructions is insufficient if there is no enemy. Saving time looks impossible.
Posted: 12/21/2012 10:07 PM
Quote
finalfighter
Experienced Forum User, Published Author, Active player (276)
Joined: 4/20/2004
Posts: 219
Location: Japan,Tokyo
NES TAS of 2020 Glitchy TAS of 2020 TAS of 2010 NES TAS of 2010 Glitchy TAS of 2010 NES TAS of 2009
It is impossible to acquire a magnet beam in the BombManStage although I did many tries. Because, when drops items will be 6x over, magnet beam can't appear. However, we need the drops item of 8x. (For the throughput which performs DelayObjectGlitch ) And your idea of HyperBomb has many objects. ( Then ,we cannot get a magnet beam) Now , our true route is BombManStage(DelayStageClear with RockBuster)->ElecMan(with HyperBomb)->IceMan(DelayFFStageClear with ThunderBeam) :)
Posted: 8/6/2012 1:12 AM
Quote
finalfighter
Experienced Forum User, Published Author, Active player (276)
Joined: 4/20/2004
Posts: 219
Location: Japan,Tokyo
NES TAS of 2020 Glitchy TAS of 2020 TAS of 2010 NES TAS of 2010 Glitchy TAS of 2010 NES TAS of 2009
I also wanted MagnetBeam, because I wants to improve entertainment elements ;) However, even if I had tried for a long time, I was not able to get it. DelayFFStageClear is a technique which can use ObjectFF at the appearing place. ObjectsFF is a special object and appears only within a specific place. ObjectsFF are actually calls to remap pattern tables of PPU. If NMI is put in the specific processing of ObjectFF, a mistake will occur. It is only a specific place of an IceManStage, BombManStage that can use DelayFFStageClear now.
Posted: 8/4/2012 9:11 PM
Quote
finalfighter
Experienced Forum User, Published Author, Active player (276)
Joined: 4/20/2004
Posts: 219
Location: Japan,Tokyo
NES TAS of 2020 Glitchy TAS of 2020 TAS of 2010 NES TAS of 2010 Glitchy TAS of 2010 NES TAS of 2009
Probably, next, I need to consider a boss's capture order. 3 plan. 1: IceMan:4xdrop item and ThuderBeam(DelayStageClear) , clear point time(50000) 2:BombMan:8xdrop item and RockBuster(DelayStageClear) , clear point time(50000) 3:BombMan:4xdrop item and ThuderBeam , change arm time(DelayStageClear) + clear point time(50000) I need to consider the course in which the change of arms does not become slow. :)
Posted: 8/4/2012 9:01 PM
Quote
finalfighter
Experienced Forum User, Published Author, Active player (276)
Joined: 4/20/2004
Posts: 219
Location: Japan,Tokyo
NES TAS of 2020 Glitchy TAS of 2020 TAS of 2010 NES TAS of 2010 Glitchy TAS of 2010 NES TAS of 2009
I challenged the proof of DelayFFStageClear in an IceMan stage similarly. I am using ThuderBeam and 4x drop item by IceManStage and proved that DelayFFStageClear is possible. :) We will be able to save about 1 min.
Posted: 8/4/2012 8:52 PM
Quote
finalfighter
Experienced Forum User, Published Author, Active player (276)
Joined: 4/20/2004
Posts: 219
Location: Japan,Tokyo
NES TAS of 2020 Glitchy TAS of 2020 TAS of 2010 NES TAS of 2010 Glitchy TAS of 2010 NES TAS of 2009
I reached in BombManStage the conclusion that acquisition of a DelayMagnetBeamAdapter was impossible. Execution of DelayObjectGlitch requires 8 x drops items, however, if an drop item increases, the kind of appearing object will be restricted. The screen end of BombManStage: use 8 x drop item and use RockBuster , appearing DelayObject=0,5,10,11 4xdrop item and use ThuderBeam , appearing DelayObject= 0-0xff This means that the acquisition by RockBuster is impossible.
1 2
8 9