This is a short Skate or Die run that is only about 2 minutes long. It attempts to beat the game as fast as possible by taking optimal routes and manipulating the AI to end some of the events as soon as possible.
Game objectives
Emulator used: BizHawk 2.3.2
Aims for fast time
Luck Manipulation
Comments
This is a short TAS that mostly involves optimal movement and inputs, except in the last two stages where you can use your character's movement to manipulate the AI's position resulting in the stage ending faster. The third stage's optimizations just include staying off grass and hitting a perfect straight line to minimize left and right movement. Other than that you just need to end the ramp stages as quickly as possible by doing nothing and pumping your character for speed.
Stage by stage comments
Stage 1 - Freestyle
To complete this stage as fast as possible you need to turn back without launching in the air. On the last pass you can pump your character (rapid press A) to gain speed. That's it. Not much going on.
Stage 2 - High Jump
In this stage you want to make the smallest jump possible then pump the character to gain speed on the return trip.
Stage 3 - Downhill
You want to stay off the grass and pick the perfect line at the skull ramp to hit most of the obstacles in a straight line. However, in the beginning it's necessary to make a sharp turn across the grass. Rocks are jumped over to avoid losing time by turning around them.
Stage 4 - Jam
This is the first stage with AI manipulation. The reason you need to manipulate the AI is because the stage ends when Lester crosses the finish line, rather than when you cross the finish line. To the best of my knowledge, Lester's movement is determined based on your character's position. For most of the level it doesn't matter but at the end of the level you want to be in a position that keeps him from lagging behind. I accomplish this by straightening my character at the end of the level during a range of acceptable frames. If you don't do this in the right spot Lester can get caught up on the speed bump or decide to turn right and bonk into something on the right side.
Stage 5 - Joust
This level also has AI manipulation which comes into play at the final hit. Again, I think the AI's actions are determined based on what player 1 does just like in the previous level. Before the winning hit you want Pete to launch up above the lip of the pool which will allow you to line up with him when the game gives you the joust pole. You set this up when you are on the left side pool with your Y position at the apex of your turn. When you roll back to the right side of the pool you want to position player 1 just a couple pixels below Pete so that he doesn't hit you when he has the pole then input up and input A to hit him before he rolls down the pool. Normally you have to both roll to the left side of the pool for the final hit.
This is a pretty short run and for a portion of it you are more or less just waiting for the stage to end. This game doesn't offer many opportunities for saving time so I had to think about how I can manipulate the opponents in the last two stages to find more time to save. Doing this I was able to save a little over 2 seconds on my original TAS where I didn't consider this.
Areas of improvement would include finding a way to make Pete cross the finish line sooner and getting the last hit in the joust sooner. Maybe a glitch could be found to end level 2 (Freestyle) before you reach the 10th pass.
So one potential complaint one could have is that parts of the game are obviously not built for being completed in a speedy manner. The goal chosen "compete all" goes through all 5 different events. However only one of those parts is trivial to optimize and so the goal seems reasonable.
The last event you are given a choice of different opponents, each labeled with a different difficulty. This one chooses to fight against the easiest opponent. Given the hardest difficulty guideline is... well... a guideline, and selecting the hardest difficulty here wouldn't make it substantially more interesting, I'd say that's it's fine to use easiest.
The TAS was not received well by the audience in terms of entertainment and for good reason. Each minigame is kinda simplistic and not particularly interesting.
Joined: 10/12/2011
Posts: 6438
Location: The land down under.
Spiked the input by 78 frames.
TAS is as boring as expected but I'll give it a Meh.
As a note: Author's encode is RTA timing which is why the times don't match up.
WebNations/Sabih wrote:
+fsvgm777 never censoring anything.
Disables Comments and Ratings for the YouTube account.Something better for yourself and also others.
This should be tagged "easiest difficulty" since you're jousting against Pete instead of Lester. And since you're luck-manipulating, I wonder if Lester can't be manipulated in the same way?
I have fond memories of ski or die. This TAS is sad.
I don't really see the point of 'trying for best time' in a game like this. It's about a superplay, not sucking as hard as you can as fast as you can.
This run is the equivalent of a tetris run where you mash the blocks down as fast as you can to lose quickly. No Vote.
This should be tagged "easiest difficulty" since you're jousting against Pete instead of Lester. And since you're luck-manipulating, I wonder if Lester can't be manipulated in the same way?
Lester waits until you are on his side of the pool to drop down and ends up behind you when you get the jousting pole on the 5th pass. The only way I found I can change his position that drastically is by stopping at the lip and waiting for him to catch up, which would make my 5 passes take longer.
So this game essentially composes of multiple events. Performing poorly in some events like the first two allow you to end them quicker. In a couple of others, completing them quickly is considered performing them well. However, no matter how well you perform, the game doesn't appear to treat it any differently. You always move on to the next round. If you lose in the final jousting event, the only difference is that the word lose appears and you get less score. You then reappear in the skate shop same as usual. According to the movie rules:
Where applicable, the movie must reach an ending screen that positively signifies a game is finished successfully. Reaching a game-over screen is not considered beating the game. If a game shows the same ending screen regardless of success or failure, reaching it is not considered successful completion.
Is this completion?
You could say this movie completes all new content according to endless games rules, but is this completion if the game does not care about how well you do in the slightest?
Individual events (Downhill and Jam) certainly can be considered non-trivial in that they have a clear goal: complete as quickly as possible. Compete all does not have such a clear goal.
[16:36:31] <Mothrayas> I have to say this argument about robot drug usage is a lot more fun than whatever else we have been doing in the past two+ hours
[16:08:10] <BenLubar> a TAS is just the limit of a segmented speedrun as the segment length approaches zero
Is this completion?
You could say this movie completes all new content according to endless games rules, but is this completion if the game does not care about how well you do in the slightest?
Individual events (Downhill and Jam) certainly can be considered non-trivial in that they have a clear goal: complete as quickly as possible. Compete all does not have such a clear goal.
Can losing an event that is only intended to be 'beaten' and isn't time based actually be considered as 'completing' it?
It's more of a 'participated-in' status instead of 'completed'.
Sure, losing may be the fastest way to show the content, but I'd argue that only losing in such a competition event doesn't really count as 'completion'.
Even if the game ultimately doesn't care how you perform in the task presented; the general assumption can be made that the game still expects you to try to succeed in that task. Thus losing intentionally isn't 'completing' the task presented.
EDIT: This would really only apply to the last stage for this game. My comment was more generalized and not specific to this game.
The first two modes of this game are done satisfactorily for a fast finish.
As presented, this final stage is good as well.
EDIT#2: Regarding completion of this game: If the two poorly performed events are the beginning aren't included because they don't have a clear completion goal, the run wouldn't show all unique content of the game. Therefore, even though the game doesn't care how poorly they are done, they are necessary for a 'full-completion' run of this game. These events, as presented, still appear optimized for time.
The only other option would be to allow multiple publications of individual events as 'complete' runs (which in my opinion, isn't full completion of the game).
Is this completion?
You could say this movie completes all new content according to endless games rules, but is this completion if the game does not care about how well you do in the slightest?
Individual events (Downhill and Jam) certainly can be considered non-trivial in that they have a clear goal: complete as quickly as possible. Compete all does not have such a clear goal.
Can losing an event that is only intended to be 'beaten' and isn't time based actually be considered as 'completing' it?
It's more of a 'participated-in' status instead of 'completed'.
Sure, losing may be the fastest way to show the content, but I'd argue that only losing in such a competition event doesn't really count as 'completion'.
Even if the game ultimately doesn't care how you perform in the task presented; the general assumption can be made that the game still expects you to try to succeed in that task. Thus losing intentionally isn't 'completing' the task presented.
EDIT: This would really only apply to the last stage for this game. My comment was more generalized and not specific to this game.
The first two modes of this game are done satisfactorily for a fast finish.
As presented, this final stage is good as well.
EDIT#2: Regarding completion of this game: If the two poorly performed events are the beginning aren't included because they don't have a clear completion goal, the run wouldn't show all unique content of the game. Therefore, even though the game doesn't care how poorly they are done, they are necessary for a 'full-completion' run of this game. These events, as presented, still appear optimized for time.
The only other option would be to allow multiple publications of individual events as 'complete' runs (which in my opinion, isn't full completion of the game).
My question essentially comes down to the section of the rules defining games which have achievable goals. Is simply ending the gameplay segments as quickly as possible the same thing as completing them? As you pointed out it's not "losing them" but it's not clearly winning them either. Typically in a video game you play it well to progress further into them but here progression happens regardless of performance. That signals to me that "progression" seems like a trivial goal.
I am not against accepting this kind of run despite feeling that a score oriented run would be more interesting, but I would like clarification on a rather ambiguous section of the rules.
[16:36:31] <Mothrayas> I have to say this argument about robot drug usage is a lot more fun than whatever else we have been doing in the past two+ hours
[16:08:10] <BenLubar> a TAS is just the limit of a segmented speedrun as the segment length approaches zero
My question essentially comes down to the section of the rules defining games which have achievable goals. Is simply ending the gameplay segments as quickly as possible the same thing as completing them? As you pointed out it's not "losing them" but it's not clearly winning them either. Typically in a video game you play it well to progress further into them but here progression happens regardless of performance. That signals to me that "progression" seems like a trivial goal.
I am not against accepting this kind of run despite feeling that a score oriented run would be more interesting, but I would like clarification on a rather ambiguous section of the rules.
This feels similar to my C64 Decathlon submission. It similarly didn't require good performance for progression. Granted it was rejected because the aim for max score didn't allow for true fastest completion (which would have been effectively done by just failing all non-timed events). That game's judgment has been suggested to be rejudged given the clarified vault rules for max-score runs (as somewhat discussed here).
Skate or Die may be a similar situation where only a max-score run can be vault eligible and there's no feasible way to have a speed only related vault eligible submission (barring perhaps ACE).
Thinking of it this way, it seems we either have 2 out of 5 events designed for high score or 3 out of 5 events designed for speed. Would the best run be one that aims for high score on the first 2 and speed on the next 3? I'm not sure how switching goals in the middle of the run would be any better than a run where 2/5ths of it is essentially waiting for the event to end.
Thinking of it this way, it seems we either have 2 out of 5 events designed for high score and 3 out of 5 events designed for speed. Would the best run be one that aims for high score on the first 2 and speed on the next 3? I'm not sure how switching goals in the middle of the run would be any better than a run where 2/5ths of it is essentially waiting for the event to end.
For a high score TAS: If speed is the only metric for score on an event, then the fastest time = the best score. If there are other ways of earning points, attempting to maximize score takes priority over speed to the end point. It could theoretically end up being a balance point between speed and other score earned in-stage that yields the maximum score.
So for Freestyle and High Jump, just max the score.
For Downhill and Jam, you'd need to maximize points by balancing speed and points from tricks, pickups, etc.
For Joust, it's probably still a 'win as fast as possible' situation.
Lester waits until you are on his side of the pool to drop down and ends up behind you when you get the jousting pole on the 5th pass. The only way I found I can change his position that drastically is by stopping at the lip and waiting for him to catch up, which would make my 5 passes take longer.
Sure, but the movie rules allow you (in fact, encourage you) to use a higher difficulty level even if it's slower. Higher difficulties are usually slower but may lead to more entertaining runs.
Joined: 4/17/2010
Posts: 11475
Location: Lake Chargoggagoggmanchauggagoggchaubunagungamaugg
Radiant wrote:
simillarian wrote:
Lester waits until you are on his side of the pool to drop down and ends up behind you when you get the jousting pole on the 5th pass. The only way I found I can change his position that drastically is by stopping at the lip and waiting for him to catch up, which would make my 5 passes take longer.
Sure, but the movie rules allow you (in fact, encourage you) to use a higher difficulty level even if it's slower. Higher difficulties are usually slower but may lead to more entertaining runs.
There's a difference between being just slower yet harder, and showcasing more impressive prowess with this harder difficulty. If it's just more of the same content, higher difficulty doesn't make much sense.
Memory wrote:
So this game essentially composes of multiple events. Performing poorly in some events like the first two allow you to end them quicker. In a couple of others, completing them quickly is considered performing them well. However, no matter how well you perform, the game doesn't appear to treat it any differently. You always move on to the next round. If you lose in the final jousting event, the only difference is that the word lose appears and you get less score. You then reappear in the skate shop same as usual. According to the movie rules:
Where applicable, the movie must reach an ending screen that positively signifies a game is finished successfully. Reaching a game-over screen is not considered beating the game. If a game shows the same ending screen regardless of success or failure, reaching it is not considered successful completion.
Is this completion?
You could say this movie completes all new content according to endless games rules, but is this completion if the game does not care about how well you do in the slightest?
Individual events (Downhill and Jam) certainly can be considered non-trivial in that they have a clear goal: complete as quickly as possible. Compete all does not have such a clear goal.
How poor does performance have to be to end those 2 first levels ASAP? How much faster does it end them compared to this movie? Regarding the last level, I'd argue that the game is indicating that you failed to complete it.
In general, if all levels are optional and some of them are too trivial to fastest-complete, they can be excluded from an any% movie that just plays the levels sensible to speedrun. This comes from the fact that having all levels as optional makes them all kinda separate modes, and those can have individual branches, yet we prefer to have an all-in-one movie for those. And some modes don't make sense in a speedrun, so they wouldn't become branches, therefore they wouldn't make sense in an all-in-one movie either.
Then there can be a branch that tries to maximize the overall score as long as there's no other way to define competitive full completion for this game, so that might play "complete all".
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.
So this game essentially composes of multiple events. Performing poorly in some events like the first two allow you to end them quicker. In a couple of others, completing them quickly is considered performing them well. However, no matter how well you perform, the game doesn't appear to treat it any differently. You always move on to the next round. If you lose in the final jousting event, the only difference is that the word lose appears and you get less score. You then reappear in the skate shop same as usual. According to the movie rules:
Where applicable, the movie must reach an ending screen that positively signifies a game is finished successfully. Reaching a game-over screen is not considered beating the game. If a game shows the same ending screen regardless of success or failure, reaching it is not considered successful completion.
Is this completion?
You could say this movie completes all new content according to endless games rules, but is this completion if the game does not care about how well you do in the slightest?
Individual events (Downhill and Jam) certainly can be considered non-trivial in that they have a clear goal: complete as quickly as possible. Compete all does not have such a clear goal.
How poor does performance have to be to end those 2 first levels ASAP? How much faster does it end them compared to this movie? Regarding the last level, I'd argue that the game is indicating that you failed to complete it.
"Worse" performance ends the first level slower. The goal of the first level is to rack up as much points as possible within 10 passes across the center of the half pipe. This movie ends this level as quickly as possible by making small jumps that don't earn much score. It's possible to bail enough times that you end the level but don't gain any score, but this is slower. The second level is completed seemingly as poorly as it is allowed (minimal height on the jump and then ends the level by jumping off their board on the other side.
In general, if all levels are optional and some of them are too trivial to fastest-complete, they can be excluded from an any% movie that just plays the levels sensible to speedrun. This comes from the fact that having all levels as optional makes them all kinda separate modes, and those can have individual branches, yet we prefer to have an all-in-one movie for those. And some modes don't make sense in a speedrun, so they wouldn't become branches, therefore they wouldn't make sense in an all-in-one movie either.
Then there can be a branch that tries to maximize the overall score as long as there's no other way to define competitive full completion for this game, so that might play "complete all".
The minigames can be accessed separately but this movie chooses to partake in the "compete all" mode which forces you to do all of them back to back.
[16:36:31] <Mothrayas> I have to say this argument about robot drug usage is a lot more fun than whatever else we have been doing in the past two+ hours
[16:08:10] <BenLubar> a TAS is just the limit of a segmented speedrun as the segment length approaches zero
Joined: 4/17/2010
Posts: 11475
Location: Lake Chargoggagoggmanchauggagoggchaubunagungamaugg
Memory wrote:
"Worse" performance ends the first level slower. The goal of the first level is to rack up as much points as possible within 10 passes across the center of the half pipe. This movie ends this level as quickly as possible by making small jumps that don't earn much score. It's possible to bail enough times that you end the level but don't gain any score, but this is slower. The second level is completed seemingly as poorly as it is allowed (minimal height on the jump and then ends the level by jumping off their board on the other side.
How trivial is it to finish those 2 levels ASAP?
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.
If there's a conflict between score and time, I'd say try a reasonable amount of time (no repeating tasks for more points), or go with Average Points per frame.
"Worse" performance ends the first level slower. The goal of the first level is to rack up as much points as possible within 10 passes across the center of the half pipe. This movie ends this level as quickly as possible by making small jumps that don't earn much score. It's possible to bail enough times that you end the level but don't gain any score, but this is slower. The second level is completed seemingly as poorly as it is allowed (minimal height on the jump and then ends the level by jumping off their board on the other side.
How trivial is it to finish those 2 levels ASAP?
First one can be done real time in the same time quite easily. The second is a bit more complex since you have to build up enough speed to get a jump and then end on the other ramp but it's not too hard to get a similar time.
[16:36:31] <Mothrayas> I have to say this argument about robot drug usage is a lot more fun than whatever else we have been doing in the past two+ hours
[16:08:10] <BenLubar> a TAS is just the limit of a segmented speedrun as the segment length approaches zero
Untested thought regarding Freestyle (don't have time to test at the moment): Would pumping the character between tricks to gain speed then do a plant trick to turn around be faster than what's presented here?
The plant tricks wouldn't jump, but may be a longer trick than these simple turns, but perhaps the added speed (if it actually speeds up the character) might make up for the longer trick and yield a shorter session.
Untested thought regarding Freestyle (don't have time to test at the moment): Would pumping the character between tricks to gain speed then do a plant trick to turn around be faster than what's presented here?
The plant tricks wouldn't jump, but may be a longer trick than these simple turns, but perhaps the added speed (if it actually speeds up the character) might make up for the longer trick and yield a shorter session.
I considered this too when I was making it. When I tested it with the foot plant trick, which is the quick turn around that hops off the edge, it was 4.55 seconds slower. The other trick that doesn't get air is the hand plant and it was even slower.
Joined: 4/17/2010
Posts: 11475
Location: Lake Chargoggagoggmanchauggagoggchaubunagungamaugg
Memory wrote:
First one can be done real time in the same time quite easily. The second is a bit more complex since you have to build up enough speed to get a jump and then end on the other ramp but it's not too hard to get a similar time.
Honestly it doesn't sound like a major problem overall. For a game with no ending, treating each level as its own mode is incredibly iffy. I think playing all the content in this case is saner than excluding a single event.
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.
Joined: 3/15/2018
Posts: 235
Location: United States
electricslide wrote:
I don't really see the point of 'trying for best time' in a game like this. It's about a superplay, not sucking as hard as you can as fast as you can.
This run is the equivalent of a tetris run where you mash the blocks down as fast as you can to lose quickly.
I got similar feedback when messing around with Pictionary a while back. Even my IRL friends, who aren't that into TASing (are anyone's IRL friends into TASing? Lol), said, "Aw man, I thought I was gonna get to watch you destroying all those minigames."
Maybe there's something to be said about going for high score in this game. Is there any event that would last forever with TAS powers?
Joined: 4/17/2010
Posts: 11475
Location: Lake Chargoggagoggmanchauggagoggchaubunagungamaugg
Even if there's such an event, infinite gameplay loops are banned from vaultable max score. Which means you can have a proper max score run that just avoids such techniques.
http://tasvideos.org/MovieRules.html#MaximumPoints
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.
Thank you to everyone who took the time to take a look at my submission and gave me feedback. I really appreciate it and can say that I learned a whole lot. I look forward to TASing more stuff in the future :D
