The idea of a Starfox 64 TAS aimed for speed just got a lot more interesting with the new RTA route. Currently the record is 24:12.
I wonder if a TAS could destroy all four hatches of the UFO on Katina before the cutscene starts. That, combined with a lot of lag reduction and luck manipulation, could really bring down the time.
Current project: Gex 3 any%
Paused: Gex 64 any%
There are no N64 emulators. Just SM64 emulators with hacky support for all the other games.
Lest we forget, there is an entire movie tag for maximum score.
Personally, I think maximum score as the primary goal and fastest time as a secondary goal is the best way to go about Starfox 64.
I would go for two categories:
- highscore-run, with speed as a secondary goal
- speed-oriented, with score as a secondary goal
I mean that's why we have the vault.
Current project: Gex 3 any%
Paused: Gex 64 any%
There are no N64 emulators. Just SM64 emulators with hacky support for all the other games.
So basically this is a WIP.
I know there is already a thread for Star Fox 64, but that thread is mostly around scoring since most people in that thread goes against a speedrunning TAS of the game. Although, since we are a few people who would think a speedrun is interesting, I am creating a new thread for that purpose only.
So, basically the Star Fox speedrunning community would enjoy some SF64 speedrun TASes. I have downloaded BizHawk and the ROM and I am currently theorizing different strategies. Although I have gotten across an obstacle: No one wants this TAS to be New Game. Basically the only difference is that NG wastes 10 minutes on cut scenes. I have read the "We provide the runs to be from a fresh file but you may have exceptions" so therefore I am asking for this to be an exception.
I would like to see if anyone else is interested in helping me out with this TAS (ideally, someone who is used to to create tool assisted speedruns already). I am completely new to TASing myself but I am a quick learner and I have a huge understanding of the game (tbh Zallard might be the only one I know that knows the game better than I do) since I have been running all the categories and held WR in most of them.
There is some research to do before creating the run, like optimal boosting throughout different parts of the levels and how to get passed corners the fastest. I might work on figuring this out while searching for a team player.
My plan is to start of with a Blue Line TAS to get the feeling for TASing the game, and then move on to Red Line -> Any% -> 100%.
Medal scores
stage: score/target notes
WIP
START
RESET when intro starts
Corneria: 200/150 189 before boss
Meteo: 200/200 189 before boss
Fortuna: 50/50 I'm not sure you can get 100 without wasting time. 6 before Star Wolf.
Sector X: 200/150 189 before boss
Titania: 200/150 189 before boss
RESET after Titania
---
Katina: 150/150 gives you enough time to destroy the enemy mothership
Solar: 100/100 89 before boss
Macbeth: 200/150 150 before boss
Bolse: 150/150 All range mode levels can be finished quicker as it's not fixed length. It's easier with Star Wolf
RESET
---
Sector Y: 200/150 181 before boss
Aquas: 200/150 189 before boss
Zoness: 300/250 289 before boss
Sector Z: 100/100 Somewhat easier with Katt
Area 6: 300/300 289 before boss (294 to be on the safe side if not TAS)
Venom: 227/200
RESET at first credits sequence.
Rinse and repeat on expert.
So a 100% TAS is already WIP? And you are planning on doing both normal and expert in it?
If speedrunning is a factor, every movement from the arwing adds up to lose time, meaning as little movement as possible would be to prefer.
I'm completely on your side regarding NG/NG+, Pottoww. Adding 10 minutes of downtime to a 20 minute TAS for no good reason, should definitely be avoided. Just be aware of the site rules, in that you need to provide a verification movie that puts the game into the desired NG+ condition.
The only problem I have with a Star Fox 64 speed TAS is the inaccuracy of the emulation. By removing almost all of the lag, you also remove almost everything that makes Star Fox 64 speedruns interesting. I'm not saying you shouldn't go for it if it's fun for you, just be aware that such a TAS would really only be seen as a placeholder for a new attempt with an accurate emulator.
Current project: Gex 3 any%
Paused: Gex 64 any%
There are no N64 emulators. Just SM64 emulators with hacky support for all the other games.
Until someone makes an N64 emulator that emulates lag correctly (CEN64 when??), I'd prefer to see a TAS of Star Fox 64 that goes for maximum score as a primary goal.
I'm completely on your side regarding NG/NG+, Pottoww. Adding 10 minutes of downtime to a 20 minute TAS for no good reason, should definitely be avoided. Just be aware of the site rules, in that you need to provide a verification movie that puts the game into the desired NG+ condition.
While I agree with this in principle, in practice you may as well optimize the verification movie to get used to the physics (the damage route in NG+ is a hell of a lot different from Power On because the clipped wing threshold is a lot lower. This is not like the proposed Phoenix Wright NG+ where the only difference is we can zoom through dialog by holding B all day long)
I'm completely on your side regarding NG/NG+, Pottoww. Adding 10 minutes of downtime to a 20 minute TAS for no good reason, should definitely be avoided. Just be aware of the site rules, in that you need to provide a verification movie that puts the game into the desired NG+ condition.
While I agree with this in principle, in practice you may as well optimize the verification movie to get used to the physics (the damage route in NG+ is a hell of a lot different from Power On because the clipped wing threshold is a lot lower. This is not like the proposed Phoenix Wright NG+ where the only difference is we can zoom through dialog by holding B all day long)
That's not what he meant. You are referring to Expert mode, which is unlocked by getting all the medals. We were only talking about the fact that cutscenes are unskippable the first time you play the game. It's just like the first playthrough in Metroid Prime.
Sorry for not being more precise.
Current project: Gex 3 any%
Paused: Gex 64 any%
There are no N64 emulators. Just SM64 emulators with hacky support for all the other games.
The thing about the lag is that we have been researching into emulators and it seems that iQue and emulators are quite like each other when it comes to speed. iQue is already accepted as a console and the TAS would therefore be a true "fastest run" since you "could" theoretically get this time with the fastest official console as well.
Therefore I see no need to wait for a slower emulator to do a TAS when this is just as accurate compared to available consoles.
Also yes, we are talking about an already finished path through the game to skip cut scenes, not Expert Mode.
The thing about the lag is that we have been researching into emulators and it seems that iQue and emulators are quite like each other when it comes to speed. iQue is already accepted as a console and the TAS would therefore be a true "fastest run" since you "could" theoretically get this time with the fastest official console as well.
That's a very good argument you're bringing up here. If Bizhawk is pretty much the equivalent of an iQue, the TAS will feel much more legit. Also, if a good emulator really existed, that had some sort of "iQue mode", TAsers would use that anyway.
Current project: Gex 3 any%
Paused: Gex 64 any%
There are no N64 emulators. Just SM64 emulators with hacky support for all the other games.
So I am currently trying to find out the best boosting pattern for a speedrun, how should I go with this? I started off with counting the frames for each attempt with a different amount of frames boosted between the beginning of the stage and visual cue about 2000 frames into the stage, but I realized this was inaccurate due to the actual fact that I am boosting differently. To be sure mathematically I would need to play the level for the amount of frames that is evenly split with all the boost frames amounts...? Which would be a huge number...?
It may also be that it is faster to boost one way throughout some of the stage and another amount through the rest to fit it in to the rail sections. Is there anyone who can help me figure out a good way to ensure my boosting through the run is optimal?
