(Link to video)
Submission Text Full Submission Page

Emulator:

Bizhawk 2.8.0 + TAStudio, using mGBA v0.9.1.

Emulator Setting:

  • Skip BIOS: FALSE
  • RTC Use Real Time: TRUE
  • RTC Initial Time: 1/1/2010 12:00:00 AM
  • Save Type: AUTO-DETECT
  • RTC: AUTO-DETECT
  • All other settings: DEFAULT

ROM CRC32:

8686436E

BIOS:

  • GBA

BIOS SHA1:

  • 300C20DF

BIOS Region:

  • Worldwide

BIOS Revision:

  • N/A

Discussion Thread

Please do visit the discussion thread, as the thread now contains enough information regarding RNG manipulation, and things you have to keep an eye on for optimizations.

Brief Game Introduction:

The game revolves around Kunio and Ryan (the 2 main protagonists) who, after knowing about Ryan's girlfriend being kidnapped, goes to River City High School to rescue her. Along the way, Kunio and Ryan encounters many gangs in their turfs, and fights them to seek out the bosses behind the nefarious plot.

Aims for:

  • No Damages
  • RNG Manipulation
  • Speed Trade-offs for Secret Events
  • Speed / Entertainment Trade-offs
  • Abuses glitches

No Damages:

Kunio-kun (protagonist) does not take any damages, especially in this Crazy difficulty run. One single hit will take off nearly 1/3 of HP (stamina + willpower) if not careful, and can waste extra unnecessary frames.

RNG Manipulation:

Unlike the previous run, RNG manipulation is achieved by intentionally allowing the game to decrement a random number generation counter, that counts from 0x270 to 0x1. After 0x1, it goes back to 0x270. This is called the "RNG counter". The lottery ticket prize generation is determined by another value associated with the RNG counter, called the "determinant". It helps to guarantee the lottery ticket prize generation to be unique, after every cycle, rather than being predetermined by the RNG counter that loops between 0x270 and 0x1.
In the previous run, RNG manipulation is used to obtain consecutive prizes using Lotto Tix (in-game lottery tickets) in order to have enough money to buy stats-boosting items, as well as manipulating the store stock refresh item listings. For this run, RNG manipulation is used to buy a higher Lotto Tix prize (3rd prize), and quick store item stock manipulation to get a special item, Skaterz.
For every couple of frames, the game calculates what random action the enemies will take, such as jumping, running, moving up/down/left/right, attacking, or running away, subtracting from the RNG counter. There is much to learn from the mechanics of the RNG counter and how it affects the CPU enemies' responses to player actions, and enemy types throughout the areas. You may see me do weird behaviors that priorizes RNG manipulation over speed, for this reason.

Speed / Entertainment Trade-offs:

This run explores some potential improvements listed from this Submission #7569, and is mainly focused on entertainment rather than speed. Here are the list of potential improvements;
  • Manipulating a lottery ticket faster. The current run takes about 15 seconds to manipulate a 3rd prize win, while ideally it'd be possible to win 2nd prize with as little delay as possible.
  • Manipulating weak enemies to run away more often, or have the cowardly Entrees gang appear more often.
  • Manipulating our AI allies to attack more often. There is an option to change their strategy on the pause menu, but they still don’t attack nearly as much as player character.
  • Learning more about Reputation, where it's stored in memory, and how much we need to recruit Ivan. Fighting dishonorably can be faster than fighting while to maintaining a good reputation.
  • Experiment with kicking weapons against grounded bosses
  • Finding a more efficient death warp after leaving Merlin’s. If found, look into recruiting Conan instead of Ivan.
What this run covers: 1) Manipulating lottery tickets faster, 2) Kicking weapons against grounded bosses, 3) Finding more efficient warps.
The first one is about manipulating lottery tickets faster. This can be done, but you would never have any idea of what potential improvements you will get, unless you can efficiently defeat the gang types in MOOSE and ROCKO areas, and get to Merlin's Secret as close to 0x1 or 0x270 as much as possible. In this run, I tried to get the maximum value possible, 0x1E3. There are cases where, even if you get the same RNG counter values, the determinant values can increment more than one time within a single frame. Therefore, it's not guaranteed you will get the fastest 3rd Prize winnings, but it's more probable to start from the extreme ends of the RNG counter.
The second one, I had the chance of trying this out on IVAN. Unfortunately, when bosses recover by jumping into the air, they have a ton of invincibility frames until they land on the ground. These invincibility frames can also trigger when they are hit by a large weapon 1 time, including being kicked towards them. This does not have any potentials for improvement.
The third one is the most promising one, and the most cursed one of all. In Merlin's Secrets, there is an item called "Skaterz", which allows you to warp to your enemies' back when you initiate an attack action. Attack action is when you do any of the combinations of buttons that will trigger a special technique, including techniques belonging to someone else when you obtained the Custom Char scroll item. However, you require to turn this technique off for the final boss fight, as you cannot guarantee to force the bosses out-of-bounds without being able to toss them behind your back, and warping away from the ideal position to pull this off.

Notable Changes

Prior to meeting REX (bluish purple shirt with pink pompadour), reputation is essential to recruiting new allies for the secret event. Reputation is no longer essential after obtaining REX. After visiting SHEL and ABBY, IVAN is guaranteed to join you.
There are some parts in the movie where I had to sacrifice time and entertainment without gaining speed (turning off Skaterz technique). This technique actually hinders me from performing the out-of-bounds glitch in the final boss fight.

Abuses Glitches:

There are some areas in the game where it's possible to finish off the bosses pretty quickly by aligning their positions beyond the wall boundaries. This will cause them to overflow their X positions, making them out of bounds, and will fail to land on the ground, falling to their deaths. I had to use this in the boss fights, the Warehouse Battle, and the Secret Event Boss Fight.

Speedrun Information:

  • This movie is an improvement over my previous run: Submission #7554
  • This run sacrificed time for entertainment value.
  • Uses the hardest difficulty.
  • Has met 2 Secret Events, per GameFAQs.
  • Has performed RNG manipulation. However, enemy actions determined by RNG is still poorly understood.
  • Has performed warping/teleportation.

Other Information:

  • I do not use any passwords in this run.
  • There are 2 places where the game has glitched physics I use in the game, to speed up the boss battles.
  • With new information, dying gives you positive reputation! Therefore, I use death to save time and reputation score, at the cost of half of my money prior to death. It goes counter to my claim in the previous submission (linked above).
  • I did not save and restart the movie in the middle. The game doesn't have a save data system. Instead, the game saves player profiles, which is useless in this TAS run.
  • I had the help from Seirea from Speedrun.com for giving inspiration to do this TAS run.
  • I also had the help from Chamale after collaborating with him of another existing run that focused more on speed.

Commentary:

I have added some differences between this run and my previous run.
The main goal of this run is to defeat bosses in a strict order: MOOSE -> ROCKO -> BLADE -> TURK -> MOJO -> THOR -> IVAN -> DRAGON TWIN -> SLICK
When starting the game, the game menu first shows up. This is where I need to quickly configure the difficulty, as well as the text speed to VERY FAST. This helps with unskippable cutscenes with a lot of dialogues. I turned off Weapons, because this can make the run harder to complete on time and lose the "No Damages" specialty.
In order to bring out the bosses, I have to defeat all of the gang members in that area. There is a bit of a delay in the movement, to allow CPU enemies spawn in on the opposite side, therefore gathering them in one place closer to each other for quick mass attacks.
I make quick short work with the gang before MOOSE shows up. To defeat MOOSE, the easiest way is to throw him down the construction zone pit. However, this time, MOOSE chose to activate his Shuriken technique, and not his usual movement pattern of moving downwards. There is small distance between my character (ALEX) and MOOSE, to pick him up and throw him down into the pit. Despite all of that, I was able to throw him down into the pit, at the cost of some frames due to his quick recovery.
The same unskippable cutscene is there. that you had to wait through. With the text settings set to VERYFAST, this is easily breezed through.
The next boss is ROCKO. Entering the factory is where I begin fighting the gang in that area. Luckily, it's the Frat Boys, so their HP is pretty low and they are all easy to deal with. When ROCKO enters, I had to guide the enemy towards the left-most side of the factory, and then initiate the glitch to make ROCKO fall through the ground by re-aligning his position. If I don't do this, the battle will take even longer to finish.
From here, I head on to the Armstrong Thru-way, to enter the Merlin's Secret shop. After obtaining 2 Lotto Tix, I head out, and start RNG manipulation. The goal is to decrement the RNG counter as fast as possible, and this means, attacking the enemy and landing hits as quickly as I can while I move around. After 2 successful 3rd Prize winnings, I head back into the shop to buy a couple of Excaliburs and a pair of Rocketeers. Excalibur allows you to increase your attack stats. Rocketeers allow you to run at the fastest possible speed. After exiting the store, I re-entered the shop to buy the special item, Skaterz, which allows you to warp around as you attack.
After that, I quickly warp towards BLADE, and jumped into the pit to trigger another RNG manipulation, when this time, I manipulate the bookstore's shop items to give me Grand Slam. Grand Slam is a technique that allows you to swing your weapon 4 times in quick succession.
The third boss is BLADE, one of the 4 Zombies. He is located in Sherman Park, which requires some backtracking. Picking up his chain weapon and using Grand Slam, I tried my best to avoid hitting him while he is on the ground to sustain my reputation score.
After BLADE, I head towards TURK, and defeat the gangs there.
When battling TURK, the strategy I use involves timing and juggling him as he recovers. If you get the timing just right, he will not jump up to recover, and you can land successive blows after blows quickly.
After dealing with TURK, I head on to start the battle with the next boss, MOJO. The same strategy used against TURK is used here.
After defeating MOJO, I run and warp straight to the Vacant Factory for THOR. And the easiest way to defeat THOR is by maneuvering him into the pit.
Then I began initiating the IVAN event, where I need to meet SHEL and ABBY. After that, I moved on to meet REX. This is where you need at least some decent amount of reputation score, for REX to join you. Not too high, but not too low.
After recruiting REX, I ran straight towards River City High, and aimed for IVAN. Just like the previous run, IVAN is the start of the boss battle where it takes a ton of time to whittle down STAMINA and WILLPOWER if I didn't boost my stats up high enough. This time, I aimed to get his Willpower stat whittled down first, then use Grand Slam to hit him where it hurts. Finally, IVAN joined my party.
Heading upstairs towards the roof of River City High, the next battle is the DRAGON TWIN, ANDY and RANDY. I used another strategy involving Grand Slam and a chain weapon, to begin juggling them into a corner. I have to turn my back towards them for 2 reasons:
1) Skaterz will interfere.
2) Chain hitbox is higher than the height of my allies, preventing the enemies from jumping/thrown away from the corner as the allies bunch up together near them.
After defeating the DRAGON TWIN, before triggering the SLICK "secret meeting", I have to turn off Skaterz technique. This costs me quite a lot of frames that I cannot preserve, but it is necessary to be able to pull off the glitch to throw enemies down the roof.
With a total of 4 allies, I can then unlock the "secret event": SLICK + OTIS + DRAGON TWIN boss battle. In order to trigger this event, I have to wait until SLICK finishes his dialogue, which is why I still didn't hit him while he is talking. Then I did what I can to fight against RNG, and abuse the glitch to realign all 4 bosses off-bounds, in order to quickly end the run as fast as possible. The moment I stopped my input, is the moment my run ends, but for style, I decided to end the movie at the end of the credits BGM.

arkiandruski: Claiming for judging.
feos: From talking to the author, this run only plays through secret events that lead to the secret bosses beaten, and it aims to beat all the secret bosses barring one that's too long to reach and too weak to care beating (making it less entertaining). And damage would not have saved time. Changing the branch to "secret bosses".
arkiandruski: replacing the movie file with a faster version provided by the author.
arkiandruski: Removed 5741 empty frames from the end and reuploaded.
arkiandruski: Changing the branch name to something that more accurately describes its goals. The big goal of the run is to reach the 4v4 Slick fight as fast as possible.
arkiandruski: There are concerns about optimization and the goal, but I can tell effort went into it. I'm accepting to Playground.

TASVideoAgent
They/Them
Moderator
Joined: 8/3/2004
Posts: 14773
Location: 127.0.0.1
Post subject: The best one of all GBA RCR videos.
teoheel
He/Him
Joined: 6/26/2022
Posts: 11
Location: Prokopyevsk
That was an awesome run. Well done! That was quick and fun at the same time. Juggling Dragon Twins was an awesome moment. PS. There was one death to save time in the run, so the "I do not use any deaths to save time" line is meaningless.
tom_mai78101
He/Him
Player (99)
Joined: 3/16/2015
Posts: 160
teoheel wrote:
That was an awesome run. Well done! That was quick and fun at the same time. Juggling Dragon Twins was an awesome moment. PS. There was one death to save time in the run, so the "I do not use any deaths to save time" line is meaningless.
Nice! I copied my previous submission, and forgot to update that. :P EDIT: Now updated. Thanks again for catching that.
Active player, Reviewer (276)
Joined: 12/14/2006
Posts: 717
What is the "Secret Events" mentioned in the submission text?
tom_mai78101
He/Him
Player (99)
Joined: 3/16/2015
Posts: 160
arkiandruski wrote:
What is the "Secret Events" mentioned in the submission text?
Secret Events mentioned are: 1. Recruiting IVAN as your ally. -> Need to defeat all Zombies (bosses in red student uniforms) -> Need to meet Shel (first time) - Girl with blonde hair appears very briefly near the edge of the right side of the screen at Oak Hill Mall. -> Need to meet Shel (second time) - Same girl appears very briefly near the edge of the left side of the screen at Flatiron Mall. -> Need to meet Abby Popov (secret event unlocked) - Girl with dark purple hair appears very briefly near the edge of the left side of the screen at Oak Hill Mall. 2. Defeat SLICK, DRAGON TWIN, and OTIS. -> Need to have 4 allies -> Need to not hit SLICK until he finishes talking.
Samsara
She/They
Expert player, Senior Judge, Site Admin (2120)
Joined: 11/13/2006
Posts: 2792
Location: Northern California
I'm having trouble figuring out why this run was submitted. It's the same category as #7569: Chamale's GBA River City Ransom EX "alternate final boss" in 10:09.75, but nearly 20 seconds slower. The strategy change, while interesting, does not make up for the time loss in terms of how we judge submissions. Is there a possibility that this strategy could be optimized further and become the fastest run for this category? If so, is that being worked on in any way?
TASvideos Admin and acting Senior Judge 💙 | Cohost
warmCabin wrote:
You shouldn't need a degree in computer science to get into this hobby.
tom_mai78101
He/Him
Player (99)
Joined: 3/16/2015
Posts: 160
Samsara wrote:
I'm having trouble figuring out why this run was submitted. It's the same category as #7569: Chamale's GBA River City Ransom EX "alternate final boss" in 10:09.75, but nearly 20 seconds slower. The strategy change, while interesting, does not make up for the time loss in terms of how we judge submissions. Is there a possibility that this strategy could be optimized further and become the fastest run for this category? If so, is that being worked on in any way?
Currently working on this one to improve it faster. But my plan involves me writing a genetic algorithm bot to help me to find a better optimal path, so I can manipulate the RNG in the shop menu items and the lotto ticket winnings RNG values to get me to that place. Right now, I'm 1700 frames at the Armstrong Thruway (where the Merlin's Secrets shop is at) and I'm earlier than this run's, and possibly even earlier than the other submission's run. However, I can't seem to get the teleporting item to appear, so this is where I'm stuck with right now. Also, this submission includes footages of the potential possibilities that the other run laid bare the questions on what paths to take. There may be potential routes unlocked if I can get the optimal RNG. I believed this has entertainment value to it, unlike the other submission, so I'm submitting this run based on entertainment.
Site Admin, Skilled player (1234)
Joined: 4/17/2010
Posts: 11251
Location: RU
tom_mai78101 wrote:
I believed this has entertainment value to it, unlike the other submission, so I'm submitting this run based on entertainment.
I watched both side by side and I can't agree that there's 0 entertainment in the other submission but non-0 here. Without knowing in advance what exactly to look for, they both feel somewhat similar to someone who doesn't know the game. What should we be looking for when comparing the entertainment value of the 2 runs?
Warning: When making decisions, I try to collect as much data as possible before actually deciding. I try to abstract away and see the principles behind real world events and people's opinions. I try to generalize them and turn into something clear and reusable. I hate depending on unpredictable and having to make lottery guesses. Any problem can be solved by systems thinking and acting.
tom_mai78101
He/Him
Player (99)
Joined: 3/16/2015
Posts: 160
feos wrote:
tom_mai78101 wrote:
I believed this has entertainment value to it, unlike the other submission, so I'm submitting this run based on entertainment.
I watched both side by side and I can't agree that there's 0 entertainment in the other submission but non-0 here. Without knowing in advance what exactly to look for, they both feel somewhat similar to someone who doesn't know the game. What should we be looking for when comparing the entertainment value of the 2 runs?
I will explain in a bit more detail regarding the focus of the "entertainment" portion, which is also mentioned in the submission page under the "Speed / Entertainment Trade-offs" section. There are 2 main points for this category: Lottery Tickets Lottery tickets, or "Lotto Tix", is a shop item you can purchase from the in-game's secret shop, called "Merlin's Secret". At any point in time under normal gameplay, the player can pause the game, enter the "Items" inventory menu, and select to use the "Lotto Tix". The game will then determine which prize you have won upon using the item. There are a total of 6 types of lottery ticket winnings, and they are all determined through an RNG calculation.
    - 1st Prize - $100,000.00 - 2nd Prize - $10,000.00 - 3rd Prize - $1,000.00 - 4th Prize - $100.00 - 5th Prize - $1.00 - No Prize - No winnings.
RNG calculations for this game is poorly understood, and it's still is to this day. The only known information about the RNG is that it updates once per a range of frames, be it 1 frame, or a couple of frames. To a normal human player without using emulators, they would have very little chance of encountering a 4th Prize, let alone a 3rd Prize. The lottery ticket winnings are based off of the determinant generated by the RNG counter, and updates whenever the RNG counter decrements by 1. I've done a bit of research on what makes lottery ticket winnings be determined, by relying on brute-forcing each and every possible generated determinants, which may or may not update per frame depending on how many NPCs there are on the screen. Meaning, if I wanted to reach a specific determinant, I need to quickly bring down the RNG counter by forcing the NPCs to make their choices for retaliation. In short, NPC manipulation is tied to lottery ticket winnings. The calculation is then determined upon using the item, by passing the determinant to a function that returns 1 of the 6 possible winnings listed above. It's pretty rare to use the Lotto Tix item and receive 3rd Prize, let alone making it useful during a TAS run to be able to obtain the convenience of new special techniques and stats-boosting items in the game. Skaterz This is the bulk of the entertaining portions of the run. Skaterz is a rare item in the secret shop, that that allows you to warp to the nearest enemy and attack them from behind. When you initiate an attack, it will immediately trigger. If the player has set the "Restock" option to "Every" (meaning, the shop listings will be updated frequently), for the item to appear in the shop menu, it requires the player to check on the shop every 8 to 9 frames. And the RNG needs to line up, so that "you can get the lottery ticket winnings with enough money to buy Skaterz". What makes this item important for entertainment, is a list of criterias that a normal player would encounter:
    1. Under normal gameplay and without consulting a guide, the player would first need to discover a hidden secret item shop. 2. After discovering the shop, and without any knowledge of the available shop items, the player needs to frequently visit the shop, in order to see it. 3. The player then needs to have at least $500.00 to be able to purchase it, upon seeing the item and knowing the price tag. 4. The player needs to grind for money to earn $500.00 by defeating many gang members and recurring bosses. Playing the Lotto Tix is unfeasible, because of the RNG calculations mentioned above.
Due to its ability to warp to nearby enemies, it has so many potentials for doing a TAS run. With various special techniques in the game, a player can quickly assault the gangs with minimal actions other than attacking. When an enemy is defeated, a new enemy is spawned on the opposite side of where the player is at, only after when the enemy's dropped money is picked up or despawned. This means, with Skaterz, there is no need to run, and thus no need to spend frames on moving from 1 end of the area to the opposite side of the area if the enemies don't rush into battle, or they are focusing on your ally / allies who are playing standoffs with each other most of the times. Not only that, viewers watching the TAS run would be entertained, for it is the nature of the TAS run to show the possibilities of "being able to warp around" and attacking foes left and right. Being able to see what future TASers can do with this ability, is what makes the run enjoyable to watch and give TASers many possible routes to tackle. Thus, considering the item's rarity and potentials, please consider the conditions of "Being able to win 3rd Prize twice in a row", "Buying Skaterz", and "Warping everywhere" all within a TAS run, as part of the submission's entertainment values.
tom_mai78101
He/Him
Player (99)
Joined: 3/16/2015
Posts: 160
tom_mai78101 wrote:
Samsara wrote:
I'm having trouble figuring out why this run was submitted. It's the same category as #7569: Chamale's GBA River City Ransom EX "alternate final boss" in 10:09.75, but nearly 20 seconds slower. The strategy change, while interesting, does not make up for the time loss in terms of how we judge submissions. Is there a possibility that this strategy could be optimized further and become the fastest run for this category? If so, is that being worked on in any way?
Currently working on this one to improve it faster.
Hello, As mentioned previously, I can finally let you know that I now have an improved version of this run, still using warps and a much faster RNG manipulation. New time: 10:07.11. Used the same ROM, same core and BIOS. User movie #637953535457794582 Link to video I shall wait for the judge to update my submission.
arkiandruski wrote:
What is the "Secret Events" mentioned in the submission text?
Requesting for my new TAS run to replace the current submission, please.
Samsara
She/They
Expert player, Senior Judge, Site Admin (2120)
Joined: 11/13/2006
Posts: 2792
Location: Northern California
So, I decided to look into the Reputation system to see what was happening with it. Total Reputation is stored in IWRAM value $0242 (Combined WRAM value $040242). This only changes on screen transitions. IWRAM $0246 (Combined WRAM $040246) holds the value you've currently gained or lost on the current screen. On every screen transition, this value is added to $0242 and resets to 0. Both of these are signed, 1 byte values.
  • Running into walls gives you 1 Rep each time you do it.
  • Defeating a boss gives you 2 Rep.
  • Dying gives you 4 Rep.
  • Clearing a field of enemies gives you 5 Rep.
  • Recruiting a character gives you 16 Rep.
  • You can gain Rep while fighting, though I don't exactly know how this works.
  • Losing Rep comes from fighting dirty (hitting enemies while they're on the ground, hitting bosses while they're still talking, etc).
Recruiting Rex, as done in this run, only requires your reputation to not be negative, i.e 0 or above. What this means is that this run has a lot of wiggle room for fighting dirty to save time. 32 points of Rep, in fact: Also, since Rep only changes on screen transitions, clearing out the gang before Rex can be done however you want, as Rep cannot change between entering the screen and Rex appearing.
TASvideos Admin and acting Senior Judge 💙 | Cohost
warmCabin wrote:
You shouldn't need a degree in computer science to get into this hobby.
tom_mai78101
He/Him
Player (99)
Joined: 3/16/2015
Posts: 160
Samsara wrote:
So, I decided to look into the Reputation system to see what was happening with it.
WOW! We finally have some new data to work with. I never knew that the reputation is split into total accumulative reputation and current area reputation. No wonder I couldn't find where the reputation is at in the RAM, because I assumed the reputation score is calculated as is and assumed it's stored in 1 place. This is definitely going to be helpful, and I'll definitely improve my run next time. And here, I thought we wouldn't be able to get a sub-10 minutes TAS run... Now I think it might be possible.
Active player, Reviewer (276)
Joined: 12/14/2006
Posts: 717
So, the more I try to look into this run, the harder time I have giving it a succinct objective goal. This run and Chamale's run before it seek to unlock a final boss fight where Slick calls on three others and you fight four people. Other than that, it basically follows an any% route, except they both go out of their way to recruit Ivan, which requires talking to multiple NPCs. Currently the run is listed as "Secret Bosses". The justification for that is unlocking the final boss fight with four enemies, but if that counts as a secret boss, shouldn't those other fights mentioned be as well? Also, there are bosses that are optional which can be seen as secret or extra, like Benny and Clyde who require going on a branch from the main path, Alex/Ryan (depending on the PC you choose, you can fight the one not chosen) which requires returning to a previous area, and Jesse who requires beating a certain boss five times. The decision to recruit Ivan is also difficult. Like the Slick boss fight, I can't really say it's the only character recruit that has hidden conditions to happen. There are some characters where you have to refuse invites in order to have a chance of recruiting them for example. Also, having Ivan is not a requirement for the Slick fight, nor do I think recruiting him is the fastest way to fulfill those conditions. As far as I can tell, the best this submission can get is to be a Playground run, unless there is something I'm not seeing here.
tom_mai78101
He/Him
Player (99)
Joined: 3/16/2015
Posts: 160
arkiandruski wrote:
So, the more I try to look into this run, the harder time I have giving it a succinct objective goal. This run and Chamale's run before it seek to unlock a final boss fight where Slick calls on three others and you fight four people. Other than that, it basically follows an any% route, except they both go out of their way to recruit Ivan, which requires talking to multiple NPCs.
Yes. REX and IVAN are both recruitable allies which are the fastest 2 allies you can get apart from RYAN (playing as ALEX+RYAN), GARY (playing as ALEX only), and RICK (playing as RYAN only). The other recruitable allies require decently high reputation score, which is not feasible to do in a Crazy difficulty speedrun. Also, just found out recently, there is another way to recruit 3 allies immediately, by defeating the Generic Dudes gang 3 times, defeat the gang of any type on the street where they spawn from, and then recruiting the Generic Dudes with CONAN as the leader, but you require RYAN not to be in your party at the start. I'll try experimenting with this a bit more now that we've got Reputation nailed down. I do not think recruiting CONAN and the Generic Dudes is feasible, because: - ALEX + RYAN / RYAN + ALEX => You cannot recruit CONAN and the gang, and you can skip GARY and RICK unavoidable cutscenes. - ALEX => You can recruit CONAN and the gang, but you will accrue additional time with attempts to defeat Generic Dudes 3 times and GARY's unavoidable cutscenes. - RYAN => You cannot recruit CONAN and the gang, and you will accrue additional time with RICK's unavoidable cutscenes. EDIT: I did the experiment, and found that recruiting CONAN takes too much time. It's unfeasible. However, you do get the opportunity to RNG the lottery ticket winnings and earn 1st Prize. Refusing their invitation to let them into your party means => More unavoidable cutscenes and dialogues, and you don't want those things taking up valuable frames. Declining IVAN doesn't mean you get to recruit them back into your party later on => Failing to satisfy SLICK+3 secret event condition. I was trying to save as many frames as possible when doing the run, but in the end, I only ended up saving about 120 frames. A single dialogue can run you past 150 frames (including dialogues before and after player accepts/declines the offer), so I cannot miss any opportunities to reduce the amount of dialogues I cannot skip.
arkiandruski wrote:
Currently the run is listed as "Secret Bosses". The justification for that is unlocking the final boss fight with four enemies, but if that counts as a secret boss, shouldn't those other fights mentioned be as well? Also, there are bosses that are optional which can be seen as secret or extra, like Benny and Clyde who require going on a branch from the main path, Alex/Ryan (depending on the PC you choose, you can fight the one not chosen) which requires returning to a previous area, and Jesse who requires beating a certain boss five times.
That "Secret Bosses" category was not me who decided upon it, as I originally wanted to be called "Secret Event", based on GameFAQs walkthroughs. I thought that "Secret Bosses" category is the consensus of many, so I left it at that. In my opinion, mine and Chamale's run both should've been categorized as "Secret Event" and not "Secret Bosses", because that SLICK+OTIS+DRAGON TWIN (or called SLICK+3) boss fight battle is just a secret event, not a "secret bosses" thing. Optional boss battles, like BENNY + CLYDE (unlocks JINNY side quest), JESSE (unlocks TURK+JESSE event), TEX (requires IVAN side quest), RYAN/ALEX warehouse battle (must not have allies for secret event), and TITUS (unlocks secret ending), only contribute to additional story elements in the game, much like side quests, that rounds out the main story and character interactions. You only battle these optional bosses if you wanted to unlock more story cutscenes, see how their special techniques perform in battle, or if you just wanted to earn reputation / money. They do not contribute to the main story, and they do not block your progress to reach SLICK. The special techniques are unlocked once you are in NG+ mode, in which you will be given access to obtain the 3 CUSTOM scrolls in Merlin's Secret. Those scrolls will allow you to be anyone whom you've met before and use their techniques. For example, if you defeated JESSE, you will then be able to turn into JESSE or copy JESSE special techniques and apply it to your ALEX character. And then you can play as ALEX with JESSE techniques, if you're familiar with how the techniques are activated. If we still wanted a category that differentiates from Chamale's run and have a more succinct goal in mind, may I suggest "warping"? I did use warps/teleportation, allowing my run to be faster than Chamale's "warpless" run, and my whole goal in this run is try to use the most optimal RNG possible to make warping feasible in a TAS run.
arkiandruski wrote:
The decision to recruit Ivan is also difficult. Like the Slick boss fight, I can't really say it's the only character recruit that has hidden conditions to happen. There are some characters where you have to refuse invites in order to have a chance of recruiting them for example. Also, having Ivan is not a requirement for the Slick fight, nor do I think recruiting him is the fastest way to fulfill those conditions.
IVAN is definitely not a requirement for the SLICK fight, but it's one of the known fastest way to satisfy the requirement of "getting a total of 4 allies in your party" for the SLICK+3 boss fight, because it is the only recruitable ally that does not require any reputation. There will be more experimenting to do, now that we have an updated information of things to research (e.g. Reputation, and conditions need to be met to recruit others into your party prior to fighting against SLICK+3). Let me know if you need anything from me. I'll do my best to answer them.
Chamale
He/Him
Player (177)
Joined: 10/20/2006
Posts: 1352
Location: Canada
arkiandruski wrote:
So, the more I try to look into this run, the harder time I have giving it a succinct objective goal. This run and Chamale's run before it seek to unlock a final boss fight where Slick calls on three others and you fight four people. Other than that, it basically follows an any% route, except they both go out of their way to recruit Ivan, which requires talking to multiple NPCs. Currently the run is listed as "Secret Bosses". The justification for that is unlocking the final boss fight with four enemies, but if that counts as a secret boss, shouldn't those other fights mentioned be as well? Also, there are bosses that are optional which can be seen as secret or extra, like Benny and Clyde who require going on a branch from the main path, Alex/Ryan (depending on the PC you choose, you can fight the one not chosen) which requires returning to a previous area, and Jesse who requires beating a certain boss five times. The decision to recruit Ivan is also difficult. Like the Slick boss fight, I can't really say it's the only character recruit that has hidden conditions to happen. There are some characters where you have to refuse invites in order to have a chance of recruiting them for example. Also, having Ivan is not a requirement for the Slick fight, nor do I think recruiting him is the fastest way to fulfill those conditions. As far as I can tell, the best this submission can get is to be a Playground run, unless there is something I'm not seeing here.
I don't agree with tom_mai78101 describing it as a "secret events" run, because there are a lot of secret events and this run doesn't do most of them. I think the alternate final boss fight against 4 enemies is worth including as a separate category, but I think that's up to the voters. Recruiting Ivan is the fastest way to get four members of the party. Gary or Rick require starting with only one character, and Conan requires defeating thirty Generic Dudes.
tom_mai78101
He/Him
Player (99)
Joined: 3/16/2015
Posts: 160
Chamale wrote:
I don't agree with tom_mai78101 describing it as a "secret events" run, because there are a lot of secret events and this run doesn't do most of them. I think the alternate final boss fight against 4 enemies is worth including as a separate category, but I think that's up to the voters. Recruiting Ivan is the fastest way to get four members of the party. Gary or Rick require starting with only one character, and Conan requires defeating thirty Generic Dudes.
Understood. Then, let's not use "Secret Events" category. @feos chose "secret bosses", do you agree with this category? If yes, then I believed this should answer @arkiandruski 's concern.
tom_mai78101
He/Him
Player (99)
Joined: 3/16/2015
Posts: 160
Just noticed the branch edit. That is actually a better name. I like it, as it is more descriptive of the run's main goal. I would also liked to know what others think about this.
Site Admin, Skilled player (1234)
Joined: 4/17/2010
Posts: 11251
Location: RU
Comparison encode (don't ask why it's vertical and why the HUD looks confusing) Link to video Judging by how non-trivial it was to figure out what the actual goal of those runs was, and lack of real audience engagement, I'm not sure if this meets Moons requirements. I just watched the video I made and the only part that was really epic was the final battle, but it was incredibly short. The rest of the run was hard to understand, because routing decisions are unclear for someone who doesn't know the game. It didn't look particularly repetitive though, so I voted Meh.
Warning: When making decisions, I try to collect as much data as possible before actually deciding. I try to abstract away and see the principles behind real world events and people's opinions. I try to generalize them and turn into something clear and reusable. I hate depending on unpredictable and having to make lottery guesses. Any problem can be solved by systems thinking and acting.
tom_mai78101
He/Him
Player (99)
Joined: 3/16/2015
Posts: 160
Some external feedback has been obtained from another place. I was told that it is up to the judges to decide whether to seek out feedback from outside of TASVideo forums. Thus far, this is the only feedback I received when they viewed the comparison video posted above.
crienne wrote:
Chamale's was way better IMO. I didn't like the teleports tom_mai used and for whatever reason I found the variety of weapons in Chamale's run to be more enjoyable than tom_mai just using the chain.
Samsara
She/They
Expert player, Senior Judge, Site Admin (2120)
Joined: 11/13/2006
Posts: 2792
Location: Northern California
I guess it's time to add my opinion here. These two reasons apply to both runs on the workbench at the moment: One, I don't find the category itself to be particularly publishable. To me, it doesn't do enough to warrant a separate branch, i.e if any% to 100% is a spectrum, then this sits far too deep into the any% side and just feels like unnecessary work for effectively the same result. It makes even less sense when the final bosses are actually defeated extremely quickly with that edge of screen bug, so all of the extra effort to spawn the other three bosses leads to the payoff of spending a few more seconds throwing them. Even with the edge bug strategy, I think this final fight would be far more interesting at the end of a more "full completion" style run, perhaps something like "all bosses" since a good number are skipped here. In a way, the category choice is akin to a run that goes out of its way to use a longer setup for a game end glitch: The extra effort feels wasted because the end result ends up being nearly identical anyway. It puts in too much work to be considered fastest completion, but not enough work to differentiate itself from fastest completion in my eyes. Two, I think neither run quite meets optimization standards. This run gains over 30 seconds from early movement optimization and RNG manipulation, but the strategy it uses is clearly far slower overall than the other run. Since the RNG manipulation comes before the change in strategy, it could easily apply to both runs, meaning that a fully optimized run should be around 9:30, or even less than that. The fights in both could use some work as well from what I'm seeing. On that note, I've put this together, which improves the Moose screen by about 4-5 seconds over both runs: Link to video User movie #637993639964772884 (slightly different from the video, though only in movement after Moose) Given the dirty hit on the boss, this would likely be more of a pure any% start, though it should apply to this category as well since you only need 0 or more Reputation by Rex, you start with 16, and you're guaranteed to earn at least 28 (7 per boss, 5 from clearing out the gang and 2 from the actual boss), not to mention potential gains just from fighting in the arcane way the game expects from you. The actual strategy to clear out the gang could definitely be improved as well, I just went for the first thing that worked for me. If anyone's wondering why Moose jumps out of the pit throw here when he doesn't on the submissions, it's because of his positioning: The workbench runs wait for the dialogue to run out and for Moose to run a bit downward before the knockdown and throw, not giving him enough time to recover. For the immediate throw, he's placed just high enough on screen to be able to air recover just in time before his defeat. This still ends up being faster than the strat in the submissions. The optimization thing is sort of an interesting quandary. This run is objectively the fastest completion of this category, meaning that by our rules it's hard to justify accepting the other run as it's not the fastest known run. However, even though this is the fastest known run, it's hard to justify accepting this as well because it's clear the other run uses the far superior strategy. Since that run came first, there aren't really any excuses for this run being strategically slower. I don't think we've ever had this situation happen with two submissions on the workbench at the same time, at least not that I've seen in my time as a Judge. I've seen it happen with submissions that aim to improve published runs but end up coming short on overall execution despite having faster times. In those cases, it's a judgement call, but we can't really do that here because neither run is published. There definitely needs to be a better way of handling situations like these. I might have to go back to the drawing board for our rules and standards and figure out something that makes sense. Overall, I think the best course of action here is to just make a new strictly any% run from the ground up, with active collaboration instead of whatever led to this situation. In terms of potential improvements off of what's already here, I have a few ideas: 1. Clearly the best possible improvement would be a 2nd prize lotto win, though that seems unlikely unless the RNG is analyzed further. 2. The ultimate strategy here appears to be a combination of what the two runs do, as the Rocketeers and Skaterz are both able to work together, giving both the incredible speed and the teleportation effect. 3. Acro Circus may be worth looking into as both an attack and a movement option, though I'll admit I just love that ability and want to see it used. Ideally, it would be obtained at the same time as Grand Slam. 4. RCREX lets you freely change several options that affect both gameplay and physics. Making full use of these would speed up the run significantly. I don't exactly know how we would handle this from a rules perspective right now: I'd personally allow it, but I'd need to run it by others for a consensus before making it official. If this happens, feel free to work off of my userfile, though please at least try to improve it instead of using it unmodified.
TASvideos Admin and acting Senior Judge 💙 | Cohost
warmCabin wrote:
You shouldn't need a degree in computer science to get into this hobby.
tom_mai78101
He/Him
Player (99)
Joined: 3/16/2015
Posts: 160
Samsara wrote:
One, I don't find the category itself to be particularly publishable. To me, it doesn't do enough to warrant a separate branch, i.e if any% to 100% is a spectrum, then this sits far too deep into the any% side and just feels like unnecessary work for effectively the same result. It makes even less sense when the final bosses are actually defeated extremely quickly with that edge of screen bug, so all of the extra effort to spawn the other three bosses leads to the payoff of spending a few more seconds throwing them. Even with the edge bug strategy, I think this final fight would be far more interesting at the end of a more "full completion" style run, perhaps something like "all bosses" since a good number are skipped here.
Samsara wrote:
Overall, I think the best course of action here is to just make a new strictly any% run from the ground up, with active collaboration instead of whatever led to this situation.
Does this mean we should change it up, so that instead of using the "wall edge collision bug/glitch" to quickly off the bosses, we should aim for a "glitchless alternate final boss%"? Or, instead of fighting the 4 final bosses, we should only fight the 1 final boss, meaning it would be purely a "glitchless any%"? By using the "wall edge collision bug/glitch" where the enemy falls through the floor, the feedback I've gotten is "the final boss fights are too quick to be considered to be entertaining." Does this actually have an impact on how TASes are judged if the runs are aiming for non-standard goals/Moons, such as the cases here where the 2 runs have the same goals?
Samsara wrote:
There definitely needs to be a better way of handling situations like these. I might have to go back to the drawing board for our rules and standards and figure out something that makes sense.
I'm glad we finally helped and contributed to the site.
Samsara wrote:
In terms of potential improvements off of what's already here, I have a few ideas: 1. Clearly the best possible improvement would be a 2nd prize lotto win, though that seems unlikely unless the RNG is analyzed further. 2. The ultimate strategy here appears to be a combination of what the two runs do, as the Rocketeers and Skaterz are both able to work together, giving both the incredible speed and the teleportation effect. 3. Acro Circus may be worth looking into as both an attack and a movement option, though I'll admit I just love that ability and want to see it used. Ideally, it would be obtained at the same time as Grand Slam. 4. RCREX lets you freely change several options that affect both gameplay and physics. Making full use of these would speed up the run significantly. I don't exactly know how we would handle this from a rules perspective right now: I'd personally allow it, but I'd need to run it by others for a consensus before making it official. If this happens, feel free to work off of my userfile, though please at least try to improve it instead of using it unmodified.
For the special techniques, I will try to look into what other possible special techniques that could work in a TAS run. Note, that there is a reputation penalty system in place where if you use the same special technique 3 times in a row, you lose 2 reputation points. I believed this is a check to make sure the player isn't spamming the most powerful special technique in the game all the time and still proceeding with the story with high reputation. Please let us (me and Chamale) know on the consensus of whether we are allowed to modify in-game parameters, like gravity and the stats multipliers. For the time being, I'll keep it by their default values. In the meantime, I got to travel for work, and won't return until October. 😢
Site Admin, Skilled player (1234)
Joined: 4/17/2010
Posts: 11251
Location: RU
Samsara wrote:
4. RCREX lets you freely change several options that affect both gameplay and physics. Making full use of these would speed up the run significantly. I don't exactly know how we would handle this from a rules perspective right now: I'd personally allow it, but I'd need to run it by others for a consensus before making it official.
If those options are there to be freely and openly edited, that looks like an intended feature. Is there any official assessment of how "normal" those options are considered by the devs?
Warning: When making decisions, I try to collect as much data as possible before actually deciding. I try to abstract away and see the principles behind real world events and people's opinions. I try to generalize them and turn into something clear and reusable. I hate depending on unpredictable and having to make lottery guesses. Any problem can be solved by systems thinking and acting.
tom_mai78101
He/Him
Player (99)
Joined: 3/16/2015
Posts: 160
feos wrote:
Samsara wrote:
4. RCREX lets you freely change several options that affect both gameplay and physics. Making full use of these would speed up the run significantly. I don't exactly know how we would handle this from a rules perspective right now: I'd personally allow it, but I'd need to run it by others for a consensus before making it official.
If those options are there to be freely and openly edited, that looks like an intended feature. Is there any official assessment of how "normal" those options are considered by the devs?
I'm basing this off of how the game allows the players to access this setting. The game has 2 types of menus, the Pre-game Settings, and the In-game Settings. Pre-game settings are settings the player can configure before the main story begins. Some of the settings seen here are disabled when you're in-game, meaning you would require starting a brand new game session if you feel like you wanted to change the settings again. In-game settings are settings the player can freely toggle and tweak to their preferences. It can be accessed at any time in the game in the Start menu, but it can only be accessed after the game session has begun, meaning you cannot access these settings from the Pre-game settings menu. The gravity and stats multipliers are accessed through the in-game settings, which means, at any point in the game, it is up to the player to choose whether to stick with the defaults, change the gravity from normal to either very low or very high, or change the stats multipliers from +0 to +2 or -2. This means, the developers of the game do intend the players to choose how strong they want their characters to be, either as the main titular characters, or created through the Custom Character Scroll. (Custom Character Scroll is a NG+ item that allows the player to customize their character, allowing the player to use NPC's special techniques, NPC's stats, or custom appearances.) I personally would say, it's pretty "normal" for those options to be used in the game at any point in time by the developers.
Site Admin, Skilled player (1234)
Joined: 4/17/2010
Posts: 11251
Location: RU
Does the manual mention it anyhow, if it's available anywhere? Overall, I'd definitely allow it for Moons, but for Standard we need to know if it's a non-cheat kinda option, something intended for normal play.
Warning: When making decisions, I try to collect as much data as possible before actually deciding. I try to abstract away and see the principles behind real world events and people's opinions. I try to generalize them and turn into something clear and reusable. I hate depending on unpredictable and having to make lottery guesses. Any problem can be solved by systems thinking and acting.
tom_mai78101
He/Him
Player (99)
Joined: 3/16/2015
Posts: 160
feos wrote:
Does the manual mention it anyhow, if it's available anywhere?
Yes, it does. The screenshot depicts the stats multipliers. EDIT: Just noticed now, it's "stats modifiers", not "stats multipliers".... Bah.......