1 2 3 4
Posted: 11/17/2019 5:07 PM
Quote
FatRatKnight
Editor, Skilled player (1203)
Joined: 9/27/2008
Posts: 1085
Let's take Banjo-Kazooie. You can get the bonus scenes by having 100 jiggies. The game does track other sorts of completion such as honeycomb pieces for max health, but the ending doesn't care for that. Therefore, getting the extra scenes is not equal to collecting everything. In the same stretch in this case, you don't have to score everything in Tin Star to get the best ending. Unless someone can convince me Tin Star is an "Endless Scoring" environment (therefore making any finite score a "Targeted Scoring" goal) rather than a "Soft Maximum" environment, I will be saying that meeting the $1,000,000 threshold is simply picking and choosing which "sort-of-100%-ish" thing to get instead of simply all of them possible. Tetris is an example of an "Endless Scoring" environment, and capping the score doesn't prevent the scoring actions from continuing, but it's an entertaining target to hit the cap quickly, so I have no problems as the TAS isn't in Vault. As "Endless Scoring" implies you can continue collecting score (if you hack the score back down from the cap), collecting all scoring actions possible would be an infinitely long run. In any case, my desire for getting those high score TASes accepted are in direct conflict with the line of thinking that the ending target overwrites the score target. That is where the disagreement lies -- In that desire. I leave it to the TASVideos community whether that desire is appropriate.
Posted: 11/17/2019 5:43 PM
Quote
feos
Site Admin, Skilled player (1255)
Joined: 4/17/2010
Posts: 11495
Location: Lake Char­gogg­a­gogg­man­chaugg­a­gogg­chau­bun­a­gung­a­maugg
NES TAS of 2011
I think it should be about maximum clarity of the definition. "Max score where there is no known score limit" is unclear by definition, and that max score can always be beaten. Completing "all X" is clear, non-arbitrary, easily verifiable, and so on and so forth. If we know there is a total of 100% items in Super Metroid and one gets all 100% of them, it has non-arbitrary fullness to it. But if there's no known limit to possible score, the score you'd be getting would be unpredictable, not proven unbeatable, and in the end it'd be some arbitrary number. Just like the final movie time. In a situation where we can clearly and predictably define fullness, we prefer that. When we can't, we allow max score that doesn't have a known limit. We can't allow "unpredictably high score" on exactly the same rights as any other full completion goal. There's arguably more relative fullness in $1.3 million than in $1 million, both giving you the best ending, but there's no absolute fullness in either if you simply count by score alone, because there may be a way to get higher score.
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.
Posted: 11/17/2019 6:33 PM
Quote
FatRatKnight
Editor, Skilled player (1203)
Joined: 9/27/2008
Posts: 1085
The time metric is unclear in whether we truly did the game as fast as possible. A fair quantity of TASes are obsoleted as proof that the time isn't done to the fullest extent. We are fine as it is generally an easy quantity to measure by the length of out input file, and speedrunning as a whole is practically defined by the time it took to clear the game. A "soft maximum" score is also unclear about whether the top was reached. So, why the phobia of a score metric? Generally the whole idea I was pushing in this thread is to allow for the possibility of obsoleting another TAS by a higher score. I can answer one facet of the mentioned phobia already -- How to handle the possibility of some new-discovered "Endless Scoring" technique that break the "Soft Maximum" environment? I see TASVideos more as a speedrun site than a score run site (with the occasional but lovely entertainment-based TASes with time not as a primary goal), so if it can be turned into a time metric where possible, it is done, and if not, a score metric is the absolute last resort, if no other form of full completion is available. Targeting $1,000,000 is not a score measure, it is a time measure, further cementing my view on what this site's goals are. As such, I've already accepted my personal ideals do not fit here. A bit defeatist, but I do not desire to disrupt a working system too much. Anyway, you know my stance. I may disagree with the policy, but my goals aren't the same as the intent behind the policy. That's what I'm getting from this discussion.
Posted: 11/17/2019 7:09 PM
Quote
feos
Site Admin, Skilled player (1255)
Joined: 4/17/2010
Posts: 11495
Location: Lake Char­gogg­a­gogg­man­chaugg­a­gogg­chau­bun­a­gung­a­maugg
NES TAS of 2011
FatRatKnight wrote:
A "soft maximum" score is also unclear about whether the top was reached. So, why the phobia of a score metric? Generally the whole idea I was pushing in this thread is to allow for the possibility of obsoleting another TAS by a higher score. I can answer one facet of the mentioned phobia already -- How to handle the possibility of some new-discovered "Endless Scoring" technique that break the "Soft Maximum" environment?
We spent a ton of time and effort on brainstorming and the best solution is implemented in the current rules. Any technique that leads to infinite gameplay loop delaying completion indefinitely and scoring indefinitely, is banned, no matter when it's found. The ambiguity phobia was explained to the best of my ability on this page: http://tasvideos.org/forum/viewtopic.php?t=19596&start=25
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.
Posted: 11/17/2019 7:38 PM
Quote
FatRatKnight
Editor, Skilled player (1203)
Joined: 9/27/2008
Posts: 1085
Perhaps, but it still doesn't explain denying score metrics when a time metric for a different "full completion" is available. I had also forgotten about the discussion, but it's still just one facet I pointed out. The other facets of the phobia isn't clear to me at the moment. Still good to have pointed out my mistake there. Currently dealing with a defeatist mood, though. I hope you can understand why thinking about the direction this thread is in.
Posted: 11/17/2019 7:42 PM
Quote
feos
Site Admin, Skilled player (1255)
Joined: 4/17/2010
Posts: 11495
Location: Lake Char­gogg­a­gogg­man­chaugg­a­gogg­chau­bun­a­gung­a­maugg
NES TAS of 2011
You mean the old posts I made don't explain why we prefer clarity (of regular full completion) to ambiguity (of max score)?
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.
Posted: 11/17/2019 9:38 PM
Quote
FatRatKnight
Editor, Skilled player (1203)
Joined: 9/27/2008
Posts: 1085
I'm not thinking straight enough to answer. Wrong mood to try and analyze the discussion. Reread my second to last post before this one, erase anything after the "I can answer one facet" part of that paragraph, and see if anything else I say stands. I prefer to focus my time elsewhere until I'm ready to form a new answer.
Posted: 11/19/2019 12:06 AM
Quote
FatRatKnight
Editor, Skilled player (1203)
Joined: 9/27/2008
Posts: 1085
I have a case of a TAS I once worked on: HyperZone. Hyperzone is basically a rail shooter of sorts. There is a strictly finite quantity of enemies to shoot, so if all of them were shot down, that would be a way to define 100% completion, correct? The part I stopped at basically had the game flood you with enemies, and it is easily possible some of them fail to spawn because there aren't enough empty enemy slots. If they don't spawn, they aren't a scoring possibility. Based on the troubles I've been having trying to TAS that next part, it may be impossible to get 100% kills, so 100% kills is invalid due to impossibility, and maximum kills is not a valid 100% Vault completion goal, so my TASing goal fails to comply with this. There is something the game awards you for reaching score thresholds, new machines to fly with better capabilities. There is a finite quantity to this, and once you get the final machine, there are no further rewards in score. If you want to tie this "hard, known maximum" as a 100% completion thing, I will tell you that meeting this score threshold in an any% goal is trivial in TAS conditions, with the only questionable detail being the machine swapping animation. Therefore, by triviality, that "final machine" goal is not a significant change from any%, and is not to be accepted as having appreciably different content. The absolute last fallback of allowing maximum score to differentiate from any% would at last catch my TASing goal. But if the "final machine" goal takes priority as a measurement of completion, then I will submit there doesn't exist a meaningful Vault 100% goal to be viably submitted. As a result, score is still an unacceptable goal in Vault conditions, despite the supposed fallback. This is the line of thinking I'm dealing with at this point. A bit confrontational. Please tell me otherwise. For that matter, I still don't want to read and analyze the discussions that took place, as now these thoughts have taken hold and driven my emotions again in the wrong direction. I just don't want to deal with the subject, but now for a different reason.
Posted: 1/17/2020 9:19 PM
Quote
Desmous
Joined: 1/16/2020
Posts: 5
Saturn best tasr
1 2 3 4