A separate category might be a good idea

but what would be the category? here it's either a bad rom or something related to the emulator. If we make another category that mean other players should be able to have the same setup and compete. removing the run makes more sense, it's not a complicated run, it doesn't require many resets anyway.

did the same, timed at 2'02". This is possibly the theoretical fastest unless maybe 2'01" is possible.

I did a very quick TAS (not optmized) and I got a time of 1:59, so 2:02 is not the fastest.

I've compared the runs and put on spreadsheet. Between SteveZero and Marceau run it's not so big difference for the challenges (skateboard, snowboard, ...) but it's like 2 seconds difference between challenges, even for the TAS.

I tried different things to transition faster but couldn't figure out anything yet with the TAS. I'm guessing there are some lag reduction or frame skipping?

https://docs.google.com/spreadsheets/d/1qcknXy81ts-qcFO7yXB-WNzjMIQMVbeOMiuegnyl8_U/edit?usp=sharing

I would say it should be about "is this run is valid?" instead of "should it be a different category". If not valid it should be rejected until we figure out what's the reason

One way to figure out sub 2 min is impossible is to do a TAS. I don't know the game at all, I can explore a bit tonight

Would be great to hear from @SteveZero as well

I would agree the music sounds a bit different on @SteveZero run.

I'm suspecting the "+" was pressed by accident when the backspace key was pressed to reset the game. But I think Bizhawk would display a higher number for the FPS ...

Later tonight I can compare this run with the game run on a real console

did a rom hack to go straight to the level select, more details here:

https://www.speedrun.com/batman_returns_sms/resources/nryk4

Also implement wrapping around level but the arrow doesn't change. Let me know if you have some ideas :)

thanks! ahah yes let's go!

just updated the rom hack so pressing pause during a level would reset back to the level select

I've made a rom hack for practicing the game, more info here: https://www.speedrun.com/The_Smurfs_SMS/resources/43xlv

Let me know if there is anything you would like for this rom hack :)

FYI it's quite common, we did the same for black belt https://www.speedrun.com/black_belt/forums/ee5v3

I was able to record on my SMS console and frame count exactly like you did.

• SMS console: 5127 (1m 25s 450ms)

7 frames slower than Fusion & BizHawk

I've posted a rom hack for enabling level select easily: Black Belkt - Level Select Rom Hack

This would be useful for practice or or people without access to save state (e.g. everdrive X3)

Let me know what you think or if there are other thing I can do :)

UPDATE (Nov 11, 2024)

• Bizhawk must be set "Tool Assisted Speedrun" profile to be accurate

Here is the video with the correct profile

~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~

TL;DR - no difference between Bizhawk & Kega Fusion. The console is 2 frames slower for 20 second game

I was curious if there is any significant difference between the real console and some famous emulator. Now that I got the console at home I was able to record it and compare.

For my test I've played the first level and hold right the all time. Then I frame count from the first frame when life bar is bright red until the first frame the player start running animation.

• SMS console = 20.587 (2 frames late)
• Bizhawk 2.9.1 = 20.553
• Fusion 3.64 = 20.554

For a 11:20 run (680s) that means the console would be 1s behind. As it is today it's fine, the top 2 runs are done on emulator, but we should take that in consideration whenever someone has a 11:21 run done on the console.

(for some reason, in the video the fusion part is one frame later and I couldn't adjust it)

I got fun to work on rom hack to practice directly on the console. I've posted 2 of them in the Resources section:

• Jurassic Park (SMS) - Practice Rom Hack
• Jurassic Park (SMS) - Bosses Practice Rom Hack

I would love to completely remove the car level, that would make thing super fast. I wanted to start the level with 1 life but it didn't work as expected, like potion and medical kit don't work.

Let me know what you think or if you have ideas of rom hack I can do.

I've done exactly the same for my run.

I agree, it doesn't help to get the best time, since playing from the OBS overlay has a bit of delay and you still need to be good at turning at the right time.

In my opinion, a runner who takes the time to learn all 12 rounds would be rewarded with a better time! So yes agree it should be allowed, I think it would be great to have an option to submit a run with "play with overlay" option

I've upload a rom hack: https://www.speedrun.com/headdysms/resources/q0mtm

Let me know if I can add something to the rom hack that would help. stuff like infinite life or remove animation.

For sure, I can start writting a guide for Any%

It's quite a short run, it's all visual cues and sometimes audio cues. The all idea is to have the right space between tom and jerry so it climbs up.

I think you can start by learning the game with this run, it doesn't involve glitches or too precise inputs.

another video for a safer strategy

It's exactly the same game engine and same developers ahah You should give a try, it's fun to speedrun

since AITD1 is older it makes sense to be lower than 80k yes

For Time Gate knight chase, which is "alone in the dark 4", we set the cycle to 80000. https://www.speedrun.com/time_gate_knights_chase?h=Any_-_Medium&rules=game&x=jdzw9n6d

We had a discussion in SpeedDemoArchive here: https://forum.speeddemosarchive.com/post/time_gate_knights_chase_26.html

Thanks @JamWalker86! I just compare with the current WR and it saves 2s.

I've done a the princess castle map glitch during my yesterday's run