(Link to video)
This submission is for Caesar's Palace for the NES, a gambling simulator that is very no-nonsense and straightforward unlike other casino games of the time (Vegas Dream, Vegas Stakes, etc.) You get different endings (basically vehicles) based on how much money you leave with, so we get the best ending which is the red Ferrari-looking sportscar.
The red sportscar requires $140,000 and we start with $1,000. The game has a multitude of different casino games: slots, video poker, blackjack, roulette, and the big spinny wheel. The RNG in this game is... lacking. It advances when you're in the main casino floor (which has bathrooms for some reason. Seriously you can click on them and hear water running. NO idea why this was a thing) but not in a game. So if you enter on a specific frame, you "lock in" the result.
Looking at the games, the only real possibilities are:
Roulette: Hit something to get some cash, then hit a 36:1 spin to get to $140k. This is very slow as the bet location is farther away, and the wheel spin animation takes a while.
Video Poker: 5 coin Royal Flush in one attempt. I don't think this is even possible in the game, let alone fast enough.
Slots: $500 slot jackpot would pay $100k, so you'd need that and something else big, and I believe this would take too long.
Blackjack: WAY too slow since we have no big odds.
That leaves our savior, the big wheel. Early on in the seeds exists one where the spins are 5:1, Caesar (40:1) back to back. The screen is small so betting is quicker with less wasted motion, and the spin animation is fast. We slap down $600 on the 5:1 to win $3000, then bet exactly enough ($3400) to win $136k more on spin two, leaving at exactly $140k.
This should be the optimal path. I ran through a bunch of seeds on Slots and Poker, but wasn't able to find anything lucrative. Slots have an early seed at the $500 one where you hit for $5k, but doing that and then going to the wheel is too slow.

arkiandruski: Claiming for judging.
arkiandruski: Alright. this looks good. Accepting as a full completion

despoa: Processing...

TASVideoAgent
They/Them
Moderator
Joined: 8/3/2004
Posts: 14772
Location: 127.0.0.1
This topic is for the purpose of discussing #7800: ShesChardcore's NES Caesars Palace "best ending" in 00:41.35
LoganTheTASer
He/They
Banned User, Experienced player (765)
Joined: 7/3/2022
Posts: 458
Location: Berkshire, UK
clever strategy there.
Post subject: Movie published
TASVideoAgent
They/Them
Moderator
Joined: 8/3/2004
Posts: 14772
Location: 127.0.0.1
This movie has been published. The posts before this message apply to the submission, and posts after this message apply to the published movie. ---- [4944] NES Caesars Palace "best ending" by ShesChardcore in 00:41.35
Experienced player (866)
Joined: 9/18/2008
Posts: 148
Location: Japan
You can achieve the best ending with fewer money. I put the conditions of each endings:
EndingCondition (dec)Condition (hex)
Bus00x0
Taxi1 ..= 48630x1 ..= 0x12FF
Green car4864 ..= 117750x1300 ..= 0x2DFF
Orange car11776 ..= 655350x2E00 ..= 0xFFFF
Limousine65536 ..= 1310710x10000 ..= 0x1FFFF
Red sports car131072 ..0x20000 ..
This ending branch is processed at routine $C72B. I put some memory addresses useful for the people who want to replicate the table above:
AddressTypeDescription
$A7u8money ($100 units)
$A8u8money ($25 units)
$A9u8money ($10 units)
$AAu8money ($5 units)
$ABu8money ($1 units)
$0621u24lecalculated money (modifying this has no effect)
$077Eu8money ($25500 units)
I read some code of this game and found that it will be difficult to manipulate the RNG without a large loss. RNG routine is at $F922.
Experienced player (866)
Joined: 9/18/2008
Posts: 148
Location: Japan
I challenged the video poker and succeeded to get a royal flush, but the prize is overflowed... 400000 mod 65536 = 6784. I have nothing to say further. For reference, I put a solver (examples/poker_royal.rs) to get a royal flush. (solution)
ShesChardcore
She/Her
Skilled player (1501)
Joined: 2/23/2022
Posts: 131
Location: MN
TaoTao wrote:
I challenged the video poker and succeeded to get a royal flush, but the prize is overflowed... 400000 mod 65536 = 6784. I have nothing to say further. For reference, I put a solver (examples/poker_royal.rs) to get a royal flush. (solution)
Thank you again for your efforts. Is it possible that a slot jackpot exists that could be solved for?
Experienced player (866)
Joined: 9/18/2008
Posts: 148
Location: Japan
ShesChardcore wrote:
Thank you again for your efforts. Is it possible that a slot jackpot exists that could be solved for?
I think it may be possible. I will read the slot machine code when I have a time.