Experienced Forum User, Published Author, Senior Judge, Site Admin, Expert player
(2120)
Joined: 11/13/2006
Posts: 2792
Location: Northern California
Xujhan wrote:
I'm estimating that it'll take me a month or so to finish the improved glitchless TAS. I can take this one down if that's the consensus, but the improvements are all just minor optimizations so they shouldn't affect the entertainment quality much one way or the other.
If the movie's not going to change much overall, I can set it to delayed until you post the improved version. If it's a major change, it may be better to cancel and resubmit.
TASvideos Admin and acting Senior Judge 💙 | CohostwarmCabin wrote:
You shouldn't need a degree in computer science to get into this hobby.
Experienced Forum User, Published Author, Senior Judge, Site Admin, Expert player
(2120)
Joined: 11/13/2006
Posts: 2792
Location: Northern California
feos wrote:
Being biased due that fact isn't a reason to disregard the yes vote, because it's meant to be subjective...
My reasoning there is explicitly due to this exact situation, and I apologize for implying it was more of a general thing. I'm not against throwing out ANY vote, yes or otherwise, due to bias, but it doesn't sit quite right with me when the person ultimately responsible for a run gives an opinion that is specifically favoring keeping that run published. It's more or less like asking the author what to do with the run, the implication of "I support the run being published because I was heavily involved in making it".
Hopefully I explained that well enough.
TASvideos Admin and acting Senior Judge 💙 | CohostwarmCabin wrote:
You shouldn't need a degree in computer science to get into this hobby.
Experienced Forum User, Published Author, Senior Judge, Site Admin, Expert player
(2120)
Joined: 11/13/2006
Posts: 2792
Location: Northern California
I suppose not, but I don't see why that matters in this case if maximum score is explicitly a Vaultable category. There's no need for an any% record for this game (it is technically any% in and of itself), this is the most you can do with it, and given the non-trivial nature of maximizing score it's definitely worthy of standing on its own as an acceptable TAS in my opinion.
TASvideos Admin and acting Senior Judge 💙 | CohostwarmCabin wrote:
You shouldn't need a degree in computer science to get into this hobby.
Experienced Forum User, Published Author, Senior Judge, Site Admin, Expert player
(2120)
Joined: 11/13/2006
Posts: 2792
Location: Northern California
For the record, in the TAS, the jousting sections takes about 20 seconds per joust (a total of 1 minute for the three of them), and the flying section is about a minute and 23 seconds, a total of 2:23 for both sections. Judging from the submission notes, the underwater boss takes about 36 more seconds on Hard, so the speed differences on Hard would have to make up for that at least. It sounds like it COULD: I'd have to see evidence, though, and since I can't seem to find any longplays of anything but the PSX version, I may end up having to play the dang thing myself.
The optimization level of the rest of the run looks to be on-point, I feel like it's acceptable until we figure out whether or not Hard would actually be faster (it's a shame the two sections in question happen at the very tail end of the TAS). I'm iffy on the VBA-Next / no BIOS thing, especially given the reasoning for it (a skip that only seemed to work once on standalone VBA that isn't even used in this TAS). It's not strictly against the rules, but a future run/improvement should be on the mGBA core with the BIOS just for consistency's sake.
TASvideos Admin and acting Senior Judge 💙 | CohostwarmCabin wrote:
You shouldn't need a degree in computer science to get into this hobby.
Experienced Forum User, Published Author, Senior Judge, Site Admin, Expert player
(2120)
Joined: 11/13/2006
Posts: 2792
Location: Northern California
I wouldn't say they're similar, no. This run feels a bit more similar to something like Touhou or Ikaruga if anything. Fixed length stages, maximum score/kills, nothing there to really optimize in terms of speed. The only difference is that Ikaruga/Touhou actually have optimizable boss fights, but I don't see that as enough of a difference to throw out the comparison completely.
Max score soccer would be explicitly not aiming for fastest completion as well, since scoring goals would inherently waste time. It may be a fixed length match, but you're artificially lengthening it every time you score. Plus, depending on the game, there might not be any possible way to improve the time. If the score maxes at 99, then you'd reach 99 and would just have to play the rest of the match. In that situation, as long as you can score 99 goals at all within the time limit of a match, you can't do it any faster, since the match length is fixed and the goal animations/in-between sections would be similarly fixed.
This movie is strictly fixed length no matter what. One minute per game, move on. Scoring more within a game doesn't make it take longer, it just increases the score, and in the case that the votes on this movie trend downward back to Vault status, I think it's perfectly acceptable.
TASvideos Admin and acting Senior Judge 💙 | CohostwarmCabin wrote:
You shouldn't need a degree in computer science to get into this hobby.
Experienced Forum User, Published Author, Senior Judge, Site Admin, Expert player
(2120)
Joined: 11/13/2006
Posts: 2792
Location: Northern California
HOLY H*CK STOP POSTINGArc wrote:
Memory wrote:
That would make sense if it was not for the fact that this is clearly aiming for score...
So? A soccer movie on a fixed timer is ok as long as someone racks up a lot of goals? Why even have a fixed timer rule then?
Arc, honest question. Have you watched this run? I don't think you have, because comparing it to a soccer game isn't accurate. This is a pretty open-and-shut case as far as I'm concerned. It's 25% sports game and 75% shooter, which I don't think actually really matters in the long run: The football minigame clearly isn't a full game of football, so I don't think it strictly falls under that category, and even if it did, it's clear with the score not being able to handle going over 99 that there's a non-trivial element to TASing it.
Even with rules in place, not every submission is going to have them be applicable. If it was as easy as "Insert rule, apply to every submission", we'd more or less just be able to automate the entire judgement process. Cases like this are the entire reason we have submission threads for discussion and Judges to make the final call.
feos wrote:
The problem with this game seems to be lack of any winning condition or state whatsoever. The max score rule requires following the existing rules on movie completeness. When there's no way to speedrun the game, then there can not be any speed record for it. This movie is a score record, but I don't think Vault has been designed for that. The Decathlon movie is both, which is why it was accepted.
I see this as an infinite game with no new content after the first loop, personally.
There is technically a "fastest completion" for this game:
TASVideoAgent wrote:
Now, I aim for the maximum score in each minigame, but since the minigames are on a fixed timer, where does the speed challenge come from? Well, points can still be scored after the timer ends so it is preferrable to accumulate those last points right as the timer ticks to '0'. Additionally, some of the minigames (Submarine, most notably) have a delayed start for luck manipulation.
[...]
I waited 21 frames before switching to Submarine and 24 before starting to start spawn four ships on the left.
[...]
I move on to the next minigame as soon as possible after the max score is reached for each. In Submarine, the final score is only on screen for one frame so it is barely perceptible.
Time is optimized whenever possible, and there are potential frame improvements through better luck manipulation. The new file with a higher score is slightly longer than the original submission file, meaning there is at least a small element of time involved. Maximum score is very clearly the defined goal, but like Decathlon, fastest completion is still the subgoal. It's a matter of frames, yes, but that makes it all the more TAS-like to me. In a way, it feels like Overlord, where there's only about 5 seconds of meaningful optimization in a 4 minute movie. It's not exact, of course, but I feel there's enough here to warrant saying it's a potentially improvable speed-oriented goal. Granted, I don't even think that matters in the first place (we don't strictly need both max score AND fastest completion), but again, that's something to have a brainstorming session on in the future.
Then again, does it even matter if this movie is getting good enough votes to potentially make it to Moons? I may have jumped the gun by talking about Vault, and I apologize for that, but if the voting continues the way it has been, "fastest completion" wouldn't matter at all.
TASvideos Admin and acting Senior Judge 💙 | CohostwarmCabin wrote:
You shouldn't need a degree in computer science to get into this hobby.
I disagree. Yes there are multiple mini-games like Decathlon. But in this movie the mini-games are all on a fixed timer, with 'fixed timer' meaning that the player is forced to play for a predetermined amount of time.
If you look closer at Decathlon, a lot of the events there also technically qualify as being on fixed timers. The races don't end until the CPU finishes, the Long Jump is explicitly stated in the submission text to always be the exact same number of frames regardless of distance, the other events all require scratching/intentionally failing once the maximum score is reached... They're not explicit TIMERS, but they're still operating on some sort of fixed time.
If sports games on a fixed timer like football, soccer, and basketball are not individually Vault eligible, I don't see why combining them into a single TAS would change anything. This TAS functions the same as four consecutive fixed-timer sports games.
Or, more accurately, a single sports game and 3 shooters, and maximum score is a perfectly valid and Vaultable goal in either case anyway.
TASvideos Admin and acting Senior Judge 💙 | CohostwarmCabin wrote:
You shouldn't need a degree in computer science to get into this hobby.
Experienced Forum User, Published Author, Senior Judge, Site Admin, Expert player
(2120)
Joined: 11/13/2006
Posts: 2792
Location: Northern California
I'm actually on the fence about EB:B in general, though that's a discussion for another time, in another thread. I see the resemblance to this (coin) case, at least, though I don't think they're related enough to let one decision warrant what we do with the other.
I think another good precedent that this (coin) case resembles is this submission. I rejected that submission because there was a faster GEG run out there, despite the two runs using different glitches to reach the end. Relevant quote:
This specifically aims for the fastest glitched time. The Nico TAS that's 18 seconds faster than this one (and that's not counting this submission ending input 5 minutes early, but I won't be pedantic about that) also aims for the fastest glitched time. For the purposes of this site, "game end glitch" refers to any way that the ending is reached far earlier than expected. This could be something as simple as holding two buttons on an elevator and landing somewhere, or crashing the game in a way that lets you manipulate your way to the ending, or just going up and down stairs a bunch. "Game end glitch" is exactly how it sounds: The run uses a glitch to end the game. Both this run and the Nico TAS use glitches to end the game. Therefore, they are both "game end glitch" runs. We have a separate ACE branch, but it is only used for playaround runs. Since this was treated as an improvement to the published run (thus, the fastest run) and not as an entertainment-based category, we have to treat it as such as well.
Of course, this also isn't the exact same situation: This isn't meant to be the fastest time for the game, and it is in fact the fastest time for this particular category. The problem there is that the published CCG run was meant to be the fastest time for the game, and there now exists a much faster strategy for doing so, so even though this is strictly an improvement to the CCG category, it is still not the fastest run of the overall GEG category, and as Memory pointed out (and as was stated earlier in this discussion), this isn't even the fastest possible GEG for Gold version as there's also a method of save glitching it, which would be slower than Crystal but much faster than CCG.
I think there's a lot of inherent problems with adhering to precedents on TASvideos: They're never quite exact, they might not always be right, depending on the situation there might be several precedents that clearly contradict each other, and they might be "obsoleted" in and of themselves through rule changes, clarifications, and even just brand new runs coming in with entirely different rulesets. Because of that, to me, it's safer to reject this, obsolete CCG as a branch altogether through the Crystal save glitch run, and potentially revisit this run in the future when things inevitably change. Otherwise, keeping this means just another precedent that could complicate things even further down the road.
...Granted, the rejection solution is ALSO a precedent that could complicate things, but it's the less complicated of the two in my opinion!
TASvideos!
TASvideos Admin and acting Senior Judge 💙 | CohostwarmCabin wrote:
You shouldn't need a degree in computer science to get into this hobby.
Experienced Forum User, Published Author, Senior Judge, Site Admin, Expert player
(2120)
Joined: 11/13/2006
Posts: 2792
Location: Northern California
WARNING: Constant screen flashing throughout the Space Gunner minigame (roughly 1:02 - 2:02 in the encode, 1:00 to 2:05 to be safe). Take caution if you're sensitive.
Game seems fairly mindless, though not enough to be considered trivial in my opinion. Human high scores look to be significantly lower than in the TAS*. It's a bit hard to watch at times (I guess I already made that clear with the warning), but overall it looks pretty Vaultable to me. Maximum score is a Vaultable category, and as far as I can tell, it follows all of the listed rules. The timed nature of the minigames seems to match that of [4090] C64 Decathlon "maximum score" by DrD2k9 in 09:35.02, especially with the extra measures taken to minimize time spent between minigames.
Voting Meh on entertainment.
*There's a 1p high score of 69 for Space Gunner (1p side of this TAS gets 68 in comparison), but the TAS uses both players and the discrepancy is almost certainly due to the second player getting far more kills.
TASvideos Admin and acting Senior Judge 💙 | CohostwarmCabin wrote:
You shouldn't need a degree in computer science to get into this hobby.
Experienced Forum User, Published Author, Senior Judge, Site Admin, Expert player
(2120)
Joined: 11/13/2006
Posts: 2792
Location: Northern California
Tested four games in TAStudio and was able to reproduce the following crash on all of them:
1. Open TAStudio, turn off auto-restore if it's on, and let it create a few hundred frames of input
2. With the movie still playing, aggressively Ctrl+Ins or Ctrl+Del to add/remove frames until it crashes
This seems to only happen in those circumstances: I haven't gotten a crash with the movie paused, or with auto-restore on, though the random nature of the crash could mean I just didn't happen to get it under those circumstances during my testing. I've gotten it within 5 or 6 frames deleted/added, other times it's taken upwards of a few hundred. Ow, my arm.
TASvideos Admin and acting Senior Judge 💙 | CohostwarmCabin wrote:
You shouldn't need a degree in computer science to get into this hobby.
Experienced Forum User, Published Author, Senior Judge, Site Admin, Expert player
(2120)
Joined: 11/13/2006
Posts: 2792
Location: Northern California
Arc wrote:
What is still unclear about what I've written?
For me, it's why you're writing it in the first place.
I'm not sure what you're aiming for, here. The site already more or less operates in the way you're describing: Keep in mind we've published sports games in this same way, long before the rules for them were established and refined, allowing them to be published. If someone managed to find a way to create that same level of art out of the SNES Test Program, there could be a place for it on the site in the same way, but as of right now that's impossible given the incredibly tight constraints of what's being presented.
I'm worried that you're arguing that anything could be published, even if it falls outside the realm of emulation or even pre-established games entirely. That sounds like a really dangerous precedent to set. There has to be some level of restriction. Even if it realistically could never become a problem due to the site's quality control standards, restriction helps keep things in check when we need to figure other things out.
Hypothetical: If someone were to, say, submit a TAS that just makes a game in Game Maker, what would that be to you? It is quite literally making a game, but is it really making a game? Is the process of making a game a game in and of itself? How would that be obsoleted? Would it be writing the exact same code, but faster? What if the code is optimized in such a way that the actual base game appears unchanged? Would a brand new game be a new category? Is the entertainment value derived from the quality of the finished game, or the process of making it?
That's why restriction ends up being important. We have... Honestly, probably upwards of a hundred thousand or more potentially publishable runs with our current ruleset, which is still being adapted and tweaked. I don't think we need to expand the field in the way you're implying for a very, very long time.
---------------
On a different note: In terms of speedrunning the SNES Test Program, which is not a way I ever expected to start a sentence, I've been thinking that there are a number of other problems with that concept as a whole.
1. As it stands, only two of the six options have any actual "gameplay" in them: Controller Test and Accessories Test. Everything else is a glorified cutscene, meaning there's no real need to even include them in the TAS in the first place. Electronics Test at least has a pass/fail condition, so at the very least someone could argue for its inclusion as a gameplay element, but I personally wouldn't since the means of passing/failing it is entirely outside of your control.
2. The only really entertaining thing about the SNES Test Program is the novelty that people are speedrunning it. It's clearly Vault, which means any run has to be either any% or 100%. This run in particular can't be considered an any% run due to the fact that it actively goes to the "unnecessary" options, but it also can't be considered a 100% run because it doesn't actually do anything in Accessories Test (i.e, skipping content).
On top of that, as far as I can tell in BizHawk, the Mouse doesn't fully work in the second port, which the program requires, leading to the first test immediately failing and the third test seemingly impossible to interact with. The SuperScope tests also require it to be plugged into the second port, which CTDs BizHawk (at least with this game and BSNES core, not sure if it works elsewhere), so there doesn't actually seem to be a way to 100% the program on BizHawk. I'd try lsnes, but that would require effort.
3. User movie #65227914692814936
This completes the two gameplay elements that are possible on BizHawk (Controller Test and Accessories Test > Mouse Test > Button Test) as fast as possible, without wasting time with the "cutscenes" that are the other tests. Would this be a valid any% run? I'd still argue no, because there isn't a valid endpoint to the test program other than Genesis Denesis What Nintendenesis your SNES works, and does not in fact Nintendon't. In the same way you wouldn't do any unnecessary gameplay in order to reach a desired result (i.e, no need to watch a cutscene to figure out where to go next if you already know where to go), there's no reason to actually do anything in the run, because the desired result is technically achieved from power-on. Hell, in terms of BizHawk, the desired result is completely consistent even if BizHawk isn't open: An emulator is going to behave the exact same way for everyone, barring changes to the code or the core, in which case it's just going to behave in the exact same, but slightly better way for everyone.
...In short, the SNES Test Program is thoroughly unpublishable in every way, as far as I'm concerned. Thank you for coming to my TEST Talk.
TASvideos Admin and acting Senior Judge 💙 | CohostwarmCabin wrote:
You shouldn't need a degree in computer science to get into this hobby.
Experienced Forum User, Published Author, Senior Judge, Site Admin, Expert player
(2120)
Joined: 11/13/2006
Posts: 2792
Location: Northern California
feeuzz22 wrote:
Edit : How do we make space for a low-glitch run now (if it ever happens) ?
Ring attack already fits the bill, IMO. It has the same sort of appeal as a no-zip run, I.E it actually spends time playing levels "normally", so I don't think a separate low-glitch/no-zip run is really necessary. I think that's the best course of action for the series as a whole, actually: Showcase no-zip movement through ring attack TASes to maximize the amount of time actually spent bouncing around at various high levels of speed.
TASvideos Admin and acting Senior Judge 💙 | CohostwarmCabin wrote:
You shouldn't need a degree in computer science to get into this hobby.
Experienced Forum User, Published Author, Senior Judge, Site Admin, Expert player
(2120)
Joined: 11/13/2006
Posts: 2792
Location: Northern California
You should've submitted this on a Sunday (a black day).
Samsara, who shares a birthday with Michael Jackson’s death day. Now I know now to remember your birthday.
I take that back. You shouldn't have submitted this at all. No vote.
(voted Yes, fantastic run, loved the little attentions to detail on the entertainment and the new strats were a treat to watch)
TASvideos Admin and acting Senior Judge 💙 | CohostwarmCabin wrote:
You shouldn't need a degree in computer science to get into this hobby.
Experienced Forum User, Published Author, Senior Judge, Site Admin, Expert player
(2120)
Joined: 11/13/2006
Posts: 2792
Location: Northern California
Sync verified (this screen is where input ends, for anyone who hasn't seen the run). Fresh install of Dolphin 5.0 stable build.
Cute run, love that Daisy/Peach are the fastest with the Flower Orbs. Is there any reason for Daisy specifically, or was it just personal preference? #TeamPeach
TASvideos Admin and acting Senior Judge 💙 | CohostwarmCabin wrote:
You shouldn't need a degree in computer science to get into this hobby.
Experienced Forum User, Published Author, Senior Judge, Site Admin, Expert player
(2120)
Joined: 11/13/2006
Posts: 2792
Location: Northern California
HI YES I AM INTERESTED
EDIT: Leaning toward Option 3 at the moment, good length for the game and I like the idea of a 2p DTC run. 25 minute Saturn sounds pretty harsh, 15 minute PSX seems okay too as long as the 3D is accessible (i.e, no analog stick support).
TASvideos Admin and acting Senior Judge 💙 | CohostwarmCabin wrote:
You shouldn't need a degree in computer science to get into this hobby.
Experienced Forum User, Published Author, Senior Judge, Site Admin, Expert player
(2120)
Joined: 11/13/2006
Posts: 2792
Location: Northern California
Link to video
Live, but it's basically studio quality audio and no crowd noise getting in the way of things (except at the end). And I fucking love Lingua Ignota.
TASvideos Admin and acting Senior Judge 💙 | CohostwarmCabin wrote:
You shouldn't need a degree in computer science to get into this hobby.
Experienced Forum User, Published Author, Senior Judge, Site Admin, Expert player
(2120)
Joined: 11/13/2006
Posts: 2792
Location: Northern California
ais523 wrote:
The only reason this category exists is to avoid a reset during save, though.
The reason this category exists is because it was the fastest known method of ACE in Gen2 at the time of the original submission. The reason the category was KEPT alongside Crystal's "save glitch" run is, from what I can tell, simply nobody saying that it should be obsoleted, and nothing more than that. As far as I'm concerned, keeping CCG was a minor mistake that needs to be fixed, not a rule-defining precedent for how other GEG branches are published.
EDIT: More specifically, it seems to have been kept based on the rule of a separate branch being entertaining enough to make it to Moons, as opposed to the actual methodology of the glitch. Given that interest/entertainment value has seemingly waned in the meantime (though I could argue it wasn't that high to begin with, but I'll leave that for another post), maybe it's less of a mistake and more of a sign of the times.
TASvideos Admin and acting Senior Judge 💙 | CohostwarmCabin wrote:
You shouldn't need a degree in computer science to get into this hobby.
Sync verified (this screen is where input ends, for anyone who hasn't seen the run). Fresh install of Dolphin 5.0 stable build.
Cute run, love that Daisy/Peach are the fastest with the Flower Orbs. Is there any reason for Daisy specifically, or was it just personal preference? #TeamPeach
