Something came up in the SM64 thread which is confusing me:
The rules state that a movie must break all existing records. However, it doesn't say exactly which records must be broken. Derakon said:
Derakon wrote:
The "breaks existing records" rule is for non-assisted play. If your run isn't faster than someone sitting in front of a TV without access to slowdown and rerecords, then it's not going to be accepted (unless it's a playaround, in which case we'll bicker about it for weeks before coming to a decision).
But in the relevant section in the rules, it says "Check places such as YouTube or nicovideo for TAS movies (or speedruns)", suggesting that TAS records must be broken. Can somebody clarify?
Furthermore, would a SM64 120-star run have to at least match all single-star records? Would a full-game run in general have to match all single-level records?
The goal is to get the overall run as short as possible. So, if you lose some time in one section to save more time in another, that's fine.
I don't really think this applies to SM64's stars, the having a run be as quick as possible would naturally mean that every single star record should be matched or broken. After all, if you're slower in one star, that's a possible improvement to the run.
...well, that's what I think.
The SM64 120 star is in a difficult position. The current 120 star run is woefully out of date, due to the amount of work currently being put into individual SM64 stars. However, while that is all well and good for the SM64 community, someone attempting to do a 120 star run would be have previously achieved stars constantly outdated while in the process of making the movie. So, then, they either continue working on a suboptimal movie, i.e. possible rejection, give up, or restart from that star.
Losing proposition all around, IMO.
Sage advice from a friend of Jim: So put your tinfoil hat back in the closet, open your eyes to the truth, and realize that the government is in fact causing austismal cancer with it's 9/11 fluoride vaccinations of your water supply.
Then that would put single-star improvements and complete runs into two different categories. The former shouldn't be used as a reason to reject a complete run submission.
If we had a 120-star run on the workbench right now that combined all of the best single-star runs from, say, two months ago, I'm sure that run would be accepted. It would be known to be quite unoptimal, but at the same time it'd be a huge improvement over what we have right now.
Of course, that would entail someone making a massive run in only two months, which I don't think is feasible.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
But in the relevant section in the rules, it says "Check places such as YouTube or nicovideo for TAS movies (or speedruns)", suggesting that TAS records must be broken. Can somebody clarify?
I understand it to mean that if your goal is to beat the game as fast as possible, then your time must be faster than anything anybody else has ever done before (either unassisted or tool-assisted, and without rule-breaking cheating, of course). If there's a run of the game at SDA or at Twin Galaxies, your run must be faster. If there's a TAS either at tasvideos.org or somewhere else (and in the latter case it doesn't break the tasvideos.orgs rules), your run must be faster (although certain exceptions can be made in some cases, eg. if your run uses the U ROM while that other TAS used the J ROM).
Furthermore, would a SM64 120-star run have to at least match all single-star records? Would a full-game run in general have to match all single-level records?
Runs in different categories are independent. A 100% run doesn't need to beat individual goals (such as individual levels) faster than a any% run, if there's a good reason for that (usually, beating a sub-goal slower is necessary to make the overall run faster because that slower sub-goal allows making later parts of the game faster).
The any% run could be referred to in order to point out deficiencies in the 100% run, though. (For example, someone could point out that the any% run uses some glitch that the submitted 100% run doesn't, and ask why that is so. If there's no good reason for the 100% to not to use the same glitch, it could be grounds for rejection because the 100% run would be suboptimal.) However, in general, the 100% run must only beat other 100% runs.
