I. On the topic of ''glitchyness'' & graphical disruption in the context of entertainment:
(1):
[quote moozooh]
abusing the X-Ray results in glitches that are for most intents and purposes "major" and "game-breaking", but just a notch short of crossing the hard definition line, weaseling into the legacy categories basically on a technicality
[/quote]
(2):
[quote moozooh]
and it [Superjumps] also breaks graphics in a way that badly interferes with perception and compounds over time
[/quote]
(3):
[quote moozooh from the Any% TAS thread]
Note that while the letter of the definition listed very particular types of abuse, its spirit has been to keep egregious abuse out of the category. Bigger and badder glitches are being found and abused while the accepted definition has remained the same. And so while this run indeed doesn't violate the accepted definition, it definitely contains glitches I consider major and would like to keep out of the category. Superjumps in particular are plain awful, and X-Ray tricks, while not necessarily harmful to entertainment, leave a bad taste (I mean, what it does to boss fights is just... boring). Thus, while I am glad that this run exists, I would not like it to replace the current one. Underflow is major enough in my book, but this takes it further imo. Even the fixed graphics version is too much for comfort, and it's not even the one to base one's decisions off of.
[/quote]
(4):
[quote moozooh]
but the overall experience is irrevocably marred by glitches evidently too powerful for a legacy category.
[/quote]
(5):
[quote moozooh]
If I go by the unadulterated emulator output, it's not as entertaining as cpadolf's, sadly. It's more entertaining in multiple places, but the overall experience is irrevocably marred by glitches evidently too powerful for a legacy category.
[/quote]
[quote Habreno]
I was being very simple since I'd already made my opinion of this technique clear in Any% where it was first introduced, and you and I both know this is likely going to Stars regardless because the majority of people on this site just enjoy seeing games broken more and more (unless it's from a version change, god forbid) so the people who don't actually enjoy it are going to get shut down regardless.
[/quote]
[quote Habreno]
EZGames69 wrote:
Just because you’re not alone with your opinion doesn’t make it any more justified
Just because you don't agree with it doesn't mean it's any less justified.
[/quote]
[quote Habreno]
To quote from the Any% thread:
Memory wrote:
lxx4xNx6xxl wrote:
You are correct I just want someone that Votes No to give a reason. It is perfectly fine to Vote No everyone has an opinion. I just feel like its someone who just hates Super Metroid and didn't even watch this TAS.
If they HAD to give an opinion why would the poll be anonymous to begin with. Also again there are very valid reasons to not like this specific Super Metroid TAS. For example they may be fans of less glitchy prior TASes.
And just to remind you, that thread had seven pages (on my posts/page count, IDK about yours) and multiple people expressing the same viewpoint. You can call me out all you want but my opinion is not alone here.
[/quote]
Back then in the Any% TAS submission thread I already said I would at some point come back to it and give a response, and I kept this issue in mind over a while, and it is about time now, aswell as it is a more fitting point in time currently than just via necro-posting, so now I'll deal with all of the above quotes at once (and if one wants, one can also lump in similar statements from the Any% TAS submission thread, which I will not list in here, however):
On the Starman Guidelines page (
http://tasvideos.org/StarmanGuidelines.html ), under the therein so-called
Qualities of starred movies section, regarding the topic of
entertainment listed in there, one can find the following 2
factors specifically for entertainment, no less nor else mentioned:
[quote Starman Guidelines]
* Starred movies should above all else have very high entertainment value that can be seen from these factors:
[...]
° Good game choice, featuring intense action, fast pace, reasonable quality for the platform.
° Amount of glitches involved, and whether they merely break the game, or change the game into something never before seen.
[/quote]
Hence my logical take from this is:
If suggestions from this page meant for starred movies display any guidelines on how a movie preferably should be like, and if one strives to abide by this for decisions made for a TAS,
then as long as these 2 listed factors do not or do estimatedly not sufficiently much contradict other factors listed in there, which expliclty would be
[quote Starman Guidelines]
° High entertainment ratings from many people, with little to no negative entertainment ratings.
° Solid movement style throughout the run, that can be called god-like from a viewer's perspective.
Good artistic choices, doing many things at once, without ruining the pace.
[/quote]
(aswell as potentially further unmentioned factors), the act of choosing to incorporate Superjumps as extension of the set of glitches meant for
turning the game into something never before see, aswell as for the purpose of enhancing the
fast pace
is under the above assumption of lack of intrinsic contradiction to be seen as strictly positive bonus on the side of factors relevant for entertainment of a movie, provided one goes by these and only these guidelines for this aspect. Furthermore, alongside the entertainment aspect, the technical aspect, and the aspect of
looking too impossible to be real, one can also find the following point in the section
Qualities of starred movies (which, mind you, is also marked as bold and as consequence makes it seem to be at least on a comparable level of importance as the previously mentioned aspects):
[quote Starman Guidelines]
Starred movies should be highly innovative and do things most players would not expect, nor even considered
[/quote].
And regarding this, it was and is our viewpoint that the introduction of Superjumps and other (frequently X-Ray related) exploits would greatly benefit towards the
innovativity aswell as
unexpectedness of various contents of the 100% movie, and at very least more so than any alternative in which some or all of these exploits would have been foregone, i.e. we thought the movie then would contain more expected and considered/anticipated choices for room traversal & routes and would contain less innovative material. For example without G-mode, one would have had (at least up to current knowledge) to enter Lower Norfair at the usual entrance and the route would almost certainly lead out of Lower Norfair at its usual exit, and Maridia would have had to be entered from a more common access point, too; and a significant amount of room movement likely would have been much closer to long known approaches for these rooms from past movies. Therefore I'm highly confident in our choice in this matter having been the most conform to the above guidelines among the known options we had, and I must say I would be surprised if the general TASing audience would have a contrary view on this matter.
Facit: This circumstance is
either rather to be seen as "user error'' than as TAS decision error,
or if anything, those that complain about it should talk to TASVideos' staff about the guidelines rather than arguing over 1 specific TAS case in which those guidelines' consequences are apparent and in use!
- - -
Specified, individualized responses:
(1): TASVideos goes by its definitions from the Movie Class Guidelines (
http://tasvideos.org/MovieClassGuidelines.html ) for the categorization of TASes, so if what you are saying, moozooh, is meant to say the TAS is close to not satisfying some of the definitions' criteria anymore, then I could imagine you might be thinking of some more or less slight variations of some of these definitions. One could try to think up variations e.g. of the
heavy glitch abuse or rather the
major skip glitch definition to obtain definition cases where Superjumps would not fulfill their criteria anymore, but I'm inclined to think that if one then would blame the Superjumps being so close to this borderline (of barely satisfying the conditions), then surely for many other games one would find parallel exploits that alongside Superjumps would also have to be forced to be in a situation where they don't anymore or barely satisfy the conditions for not counting as major skip glitch (with slightly changed definition).
(2): Yes, Superjumps contribute to graphical disruption, but the major part of the problem stems from doors being entered before the screen could catch up, which (as entertainment trade-off) could have been done in a way to let the screen adapt if one chose to do so, but since there was the option to make a version with qualitatively almost entirely fixed graphics (with door tubes being an exception), it appeared to be the better choice in the end, and frame optimization beats entertainment in priority anyway, as obsoletion chains in general clearly show, with the only exception to the rule being when an author can convincingly claim to have known about a frame-improvement and chose otherwise, so doing it this way is safer.
(3),(4): I'm not sure why you are repeatedly calling the SM 100% branch a ''legacy category'', moozooh, when it according to the
Movie Class Guidelines (
http://tasvideos.org/MovieClassGuidelines.html ) is just a normal and frequently occuring
real-time optimized 100% completion category movie with an attributed set of so-called
time-saving techniques usage/avoidance consisting of
heavy luck manipulation,
heavy glitch abuse,
takes damage to save time, and
foregoes major skip glitch, turning this movie into a that way classified movie according to TASVideos' definitions; instead of e.g. being chosen to be the same except substituting the attributed
heavy glitch abuse with
foregoes time-saving glitches, and if you would like to see an optimized
real-time optimized 100% completion category movie with
heavy luck manipulation,
foregoes time-saving glitches,
takes damage to save time, and
foregoes major skip glitch attributed to it, then I don't know if or when such a movie might be made, nor to what extent the foregoing of certain glitches would be applied in it, but I'd estimate that a future in which one would go this path, there likely would a lot of debating occur when TASes start beating each other by loosening up the restrictions that correspond to the
foregoing of time-saving glitches, and might even end up running back to the current situation eventually at which we are now.
Regarding the
major skip glitch definition,
[quote Movie Class Guidelines]
Most of the time more than half of the game is skipped, compared to the fastest movie that avoids this technique.
[/quote]
, according to Sniq, without the Superjumps, the TAS would finish at a time around 1:04 or 1:05, but would still beat the previous 100% TAS.
Most of the time means at least
half of the time and at most
at all times, and it refers to either
the new movie or to
the previous movie that it is compared to. And it makes little sense to measure the amount of time that a new TAS spends in sections that it skips due to the nature of a skip spending little to no time in those sections, whereas measuring in the previous TAS how much time it spends in skipped sections does make more sense. So if one qualifies superjumping through a room as ''skipping part, or let's say even all of the game (for the section affected by a Superjump)'', then in this context, one is meant to take a current fastest movie and check for all time periods in there that correspond to (with Superjumps) skipped sections in the new TAS and then add up those time periods from the previous TAS and compare this accumulated time that corresponds to skipped sections with the previous TAS's total time to see if it is at least 1/2 of it, is how I understand this definition. Now, half of Cpadolf's TAS in 1:08:15.74 would be about 34:07 which according to Sniq's estimate on time saved due to Superjumps (of which there are in total 23 in use, by the way) would be a large multiple of about 8.5 to 11 times the actual amount of time (namely 3 to 4 minutes, going by 1:08 - 1:05 = 3 min, and 1:08 - 1:04 = 4 min) that is skipped from the previous TAS (since 8.5*4=34, and 11*3=33). So maybe it is just the fascination coming with the Superjumps, aswell to parts the difficulty of estimating how much of a difference they actually make together, that is the source of the resulting big discrepancy between the perceived influence and the actual influence, and I doubt a further 100% TAS using Superjumps could apply them in about 10 times larger magnitude of effect to get closer to this borderline set by the definition of
major skip glitch.
(5): Regarding this, it made me wonder/curious what in particular the "Did you find this movie entertaining?" question refers to/means, and since I don't think there's further specification on this on any page on TASVideos, I'd want to ask if this question is meant as
° ''overall/summed up entertainment'' (alongside the stance described by "Did you find this movie entertaining overall/on average?"), or as
° ''there exist entertaining parts in this TAS, but this might not apply to all of it'' (as in ''Did you find this movie entertaining in parts of it?")?
Which of the 2 is it (or maybe it is meant in a third, other way)? Considering moozooh's statement, if it were the 2nd case, then that'd be a ''yes'' from him, and in an overall sense, as he stated, a ''meh''. So I assume the question might be meant in the sense of the 1st case, going by conformity with moozooh's choice.
Regarding what seems to be something that worries moozooh, namely the TAS being further away from what an RTA run looks like, I think that is part of what TASes on TASVideos should even be like and what one should strive for, according to the guidelines for starred movies on the Stars page (
http://tasvideos.org/Stars.html ):
[quote Stars]
Starred movies are the premium/popular/unique movies on this site, of incredibly high quality and entertainment value.[/quote]
[quote Stars]Aims to fulfill the site's goal of exposing as many people as possible to Tool-assisted Speedrun/Superplay movies as an art form.
[/quote]
[quote Stars]
Shining example of entertaining the audience, being impressive, showing a mastery of TAS techniques, using unexpected/seemingly impossible ideas, expertise of the game.
Game choice must be one that allows for a variety of TAS techniques, strategies, and ideas. In other words, it must be difficult to make an optimized TAS.
[/quote]
Especially since moozooh was a Starman himself (to my knowledge), he should be well aware of these factors (and maybe shouldn't cling as much to nostalgic times and ways in which the game was TASed before).
- - -
II. On the topic of optimization:
[quote moozooh]
the run was submitted with a relatively large known improvement at MB2 that harkens back to the pre-TAStudio era
[/quote]
From the Low% submission text:
[quote Submission #5291]
Tourian
* Further PB lag reduction in Metroid rooms and improving their death positions while still receiving the appropriate drops.
* Further lag reduction of all phases of MB?
* Up to 75-77 frames by reducing MB2 rainbow beam timer as much as possible.
° Manipulation for this would have to begin at some point before redbeam phase.
[/quote]
For the Low% TAS submission (
http://tasvideos.org/forum/viewtopic.php?t=18579&postdays=0&postorder=asc&start=75 ), moozooh didn't complain at all about MB2. So it is somewhat surprising and out of the blue that with about 17 remaining frames that might or might not be possible to be reduced entirely (compared to the 75-77 in Low%) now make him decide this way (even though the Low% TAS was also already part of the "Tastudio era" and was made that way, except with less chances of finding someone to deeply look into MB2's behavior back then), which seems somewhat inconsistent, but maybe back then for the Low% TAS with it cutting 1% off, it was more clear and in a sovereign position compared to the previous 2 SM Low% TASes than how the situation is now with 100% and Cpadolf's work with which the new 100% TAS is compared.
Edit: Additionally, according to Sniq, Cpadolf's MB2 fight was/is improvable by about 30 to 40 frames and is aswell from the ''Tastudio era'', but Cpadolf chose to not use an input editor while TASing with Bizhawk, while lsnes existed since 2011 with input editor by that time already, and moozooh didn't complain about it at that time, but on the other hand side, the improvability wasn't mentioned in the submission text or known either at the point.
- - -
III. Sidenotes:
[quote moozooh]
superjumping past the obstacles or traversing them offscreen makes me feel cheated out of tasting TASVideos's finest wine.
[/quote]
I have a hunch that you might be criticizing Superjumps too much possibly for their perceived simplicity in thinking they trivialize rooms too much, I think, and if that is part of the reason for your feeling cheated, then maybe Sniq or I should elaborate on the various hard to notice details in which the optimization of Superjumps themselves isn't an easy task at all. Also, horizontal X-Ray-Sparks are not Superjumps either (they don't have the speed bonus and are just normal sparks except of the spark crash animation taken out), so the amount of Superjumps isn't quite as overly high.
Other than that, for the record, from 00:00 to 19:36 into the TAS, there is no broken graphics, and from 19:36 on to the end at 1:01:47 of the TAS broken graphics occur, so the relative amount is bit more than 2/3 of the total TAS time.
Considering that moozooh stated to have voted with a ''meh'', I'd like to point out that (as I have elaborated elsewhere in the past) under the assumption that the TAS will be accepted, this choice is worse in effect than taking the 4th, invisible option of not voting at all (independently of giving feedback on the TAS via posts, anyway).
[quote moozooh]
but there's no arguing it [G-Mode] makes the routing far more convenient.
[/quote]
Well, the process of routing 100% initially got harder due to its existence, since there being more options to be considered, especially with the Red Brinstar gate skip versus going down Red Tower and then up through the gGlasstube instead, and if one couldn't enter Lower Norfair from both sides, then the routing options for that would have been more restrictive, but yes, once compared to drastically different routes it stood out as favourable, and as bonus (which one can justifiedly label ''convenient''), going to Lower Norfair reversely allowed for the 2 quick elevator Superjumps (which otherwise wouldn't have worked in both cases but only once).
[quote dekutony]
Wow. You guys are so close to achieve the sub hour mark... Hopefully you'll be able to accomplish that goal one day.
[/quote]
Well, at least if one would use the timing method that is used for unassisted, normal speedruns, the TAS by that timing would be shorter than an hour, so a related goal has been achieved with this TAS.
[quote Nicos]
BUT
the xray strats are very gray area, and while you technically stay inbounds of the MAP; you don't stay inbounds of the rooms themselves;
maybe adding "of the map" on the "Forgoes out of bounds" would apease a bit the dilema
[/quote]
For inbounds questions, I suggest reading up on the matter here, but there shouldn't be any issue on this side regarding the 100% TAS: (
http://tasvideos.org/forum/viewtopic.php?p=433442&highlight=#433442 ).
- - -
Remark: Since I've in the past been told to be hard to understand at times, I introduced
bold and
italic writing styles (aswell as an attempt to structure my post further) this time for the purpose of making my points more clear to the reader.