Experienced Forum User, Published Author, Senior Judge, Site Admin, Expert player
(2121)
Joined: 11/13/2006
Posts: 2793
Location: Northern California
While that sounds like an interesting TAS-only goal, it would push the movie into nonstandard goal choice territory, I.E it would have to make Moons, and I don't see 10-15 minutes of a single boss fight making it there. Repetition is explicitly one of the reasons to choose an easier difficulty as well, though I wouldn't call this a difficulty choice so much as just intentionally making the run quite a bit slower.
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
(2121)
Joined: 11/13/2006
Posts: 2793
Location: Northern California
Spikestuff wrote:
This is faster than the RTA run when timing is compared... there's still some glaring issues which could be improved upon.
RTA timing is from starting a new game to the last hit on the last boss. I truncated the file down to roughly where the RTA record ends. It takes about 7 seconds for the TAS to reach the RTA start point, so in RTA timing this run is somewhere around 18:05, compared to the RTA record of 19:00. Spike's fix is likely accurate for TAS timing though, as the game requires further input to reach the credits.
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
(2121)
Joined: 11/13/2006
Posts: 2793
Location: Northern California
These are wonderful, thanks for sharing them! Overlord's New Mansion was a treat as someone who enjoys both highly technical-looking TASes and bears. Never knew how much I wanted to see those two things together until now.
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
(2121)
Joined: 11/13/2006
Posts: 2793
Location: Northern California
TASVideoAgent wrote:
Cardboard for writing incredibly weird stuff in his previous run's movie file's comments. That motivated me to beat it so nobody would ever have to see them ever again.
Context for this.
Well, naturally, I've been following the run since the start, and... Yes vote a thousand times over. Everything looks incredibly smooth, even rooms that are only a frame or two ahead of the published run look so much quicker just due to the movement looking so much nicer overall.
Also, haha, gaaaaaaaaaaaaaaaaaaay. <3
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
(2121)
Joined: 11/13/2006
Posts: 2793
Location: Northern California
Here's what I imagine the full workflow for a publication to be, using my own judging as a base. I can't speak accurately for the publishing part, so someone more knowledgeable please correct me where I'm wrong:
1. A submission comes in
2. A Judge claims the submission
3. The Judge watches the submission in emulator to verify sync
4. The Judge rewatches the submission more closely for optimization issues
5. If necessary, the Judge reaches out to the submitter to ask questions
6. If necessary, the Judge waits for audience feedback to come in
7. The Judge accepts the movie
8. A Publisher claims the movie
9. The Publisher dumps the full movie from power on to the desired endpoint (usually post-credits, one full loop of the ending music, but every game is different)
10. Samsara realizes she has to summarize all of this and comically wipes an anime sweatdrop from her brow while saying "Hoo boy!"
11. Depending on the game (again, every game is different), the Publisher adjusts an AviSynth file to ensure that the entire run is captured and presented accurately
11a. This may include dealing with upwards of 100 or even more distinct AVI files if the game is prone to resolution switching or if it's just really quite incredibly long, in which case these AVI files need to be stitched together. If the resolutions differ, they need to be adjusted so they're all the same resolution, otherwise they can't be encoded
11b. Three distinct encodes need to be made as well, two downloadables at console resolution and an HD streamable encode for YouTube that I believe is at least 4k and occasionally goes up to 8k
12. Needless to say, this takes a LOT of time, and Publishers have always been pretty understaffed, so they're working pretty much constantly to keep the workbench moving
13. Of course, the encoding isn't the only part. After the several hour/day encoding process (JUST the encoding, for the record, not counting the steps taken leading up to it), the encodes need to be uploaded to the proper areas (archive.org for downloadables, YT for HD), the publication screenshot needs to be taken and optimized (and occasionally chosen by the Publisher themselves if there's nothing recommended), and everything needs to be pushed to the site once it's all ready and processed
This is a several day process all-in-all, and that's not only per submission, but it could potentially be per improvement to a submission. Every time a new improvement file is posted, the workflow resets back to #3 from wherever it's at, as the Judge has to re-verify sync and re-judge the entire run to ensure that the improvement is valid. In cases where the publication process has already started, a new improvement could invalidate days of work. In cases like this, where the previous run was still on the front page before the new improvement came in, that's essentially invalidating all of the work done by the staff on the previous run.
EZGames69 wrote:
It isn’t fair to staff members here to keep having them work through it every single time. I personally try to keep my improvements for certain games spaced out, both because more improvements may be discovered and because it doesn’t stress out staff members.
This is exactly my attitude towards it, both as a TASer and as a Judge. You want to make absolutely sure there's no more improvements before working on and submitting a new run, as it helps the staff immensely, particularly the Publishers, who as I mentioned are already understaffed and overworked enough as it is, and you want to spend as much time as possible working on these improvements yourself to ensure you gain as much time from them as possible. This could mean anything from optimizing new strategies to applying them all throughout the run.
Spacing out submissions and improvements also helps the Judges as well, or at least it helps me, as I feel more confident knowing a run has a lot of time put into it, and I feel less obliged to spend several days re-TASing parts of a game to see if there are other improvements that can be slotted in. It even helps the audience, as it's clear Oddworld has seen a lot of feedback fatigue as of late due to the sheer number of new submissions and miniscule improvements over the years.
In short... Don't rush. Not every improvement has to be implemented and submitted as it's found. Build up a stockpile of improvements, re-analyze the entire run to see if there are other places where these improvements can help save time, and submit a few months down the line when you have a sizable time improvement. The audience will appreciate having noticeable improvements to watch out for, the Judges will appreciate not needing to constantly re-watch, re-check and re-judge 6 improvements over the course of a month, and the Publishers will appreciate not having to interrupt and restart their time-consuming workflow.
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
(2121)
Joined: 11/13/2006
Posts: 2793
Location: Northern California
Link to video
10 levels of Ottifanten. For the record, this submission completes 31 levels in the same amount of time.
You can pseudo-move diagonally by alternating directions, but the slower speed and different level layouts mean the solutions still need to be adjusted. Also, I feel my soul dying every minute I spend with this game, so hopefully this is enough.
EDIT: User movie #65990142120952176
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
(2121)
Joined: 11/13/2006
Posts: 2793
Location: Northern California
feos wrote:
Does anyone know how many level layouts are different between this version and the official release? We have this rule for prototypes:
Movie Rules wrote:
Prototype game versions are only accepted if there is no stable release version available.
But if more than 50% of levels are different, then those can count as different modes and both be vaultable.
I believe Ollifanten has about 30 more levels added in to begin with, but from the batch I tested (1-50), most of the level layouts are identical, albeit with some minor changes here and there.
This version is probably better, but I can't figure out how to sensibly incorporate such an exception into the above rule. If gameplay is radically different, it's feasible to compare and measure the differences, but if it's slightly different, any superiority would be mostly subjective, and hard to measure for everyone. I'd argue that whether gameplay is better or not shouldn't be a reason to prefer the prototype version in Vault, because Moons is the tier where gameplay is meant to shine and please people's tastes, and Vault is for clear cuts and measurable objective things that are true for everyone.
From my testing, the gameplay is different enough that even with identical level layouts, the "solutions" will be completely different due to Ollifanten having no diagonal movement and slower movement speed. I could do a more in-depth TAS test if need be, but I'd estimate a run of that game being over twice as long as this one is, likely even upwards of 20 minutes.
...Yeah, I'll do that now, actually. Gimme a bit of time and I'll come back with an Ollifanten test for a much better comparison.
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
(2121)
Joined: 11/13/2006
Posts: 2793
Location: Northern California
Yeah, a game not having a publication just means nobody's submitted a run that got published (nobody's submitted Equinox at all, in fact). A game doesn't have to be published in order to submit a TAS of it, so go right ahead and make/submit the run.
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
(2121)
Joined: 11/13/2006
Posts: 2793
Location: Northern California
User movie #65083822042001624
Someone else has been working on this game recently, so I decided to take a look at their userfile in comparison to this submission. According to the userfile, 4-3 is the last optimized stage in it, with everything after that being WIP input. By the end of 4-3, the userfile is 68 frames faster than the submission (this does not include BizHawk's 1 frame loading advantage), but 4-4 is also 5 frames faster, for a total of at least 73. Improvements look to be mostly movement based, particularly in walljumps, but there's also a little more lag reduction all throughout.
EDIT: I'm not entirely sure how "legit" these improvements are due to the boss in 5-2 being significantly slower, I imagine it's a global timer thing so it might be possible to salvage but I'd need to hear from someone much more familiar with the game to be sure.
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
(2121)
Joined: 11/13/2006
Posts: 2793
Location: Northern California
Oh. Uh... [quietly shoves her own recently started Castlevania Legends project under a nearby rug] Hi! Welcome to the site!
Here's the relevant section in the rules:
The movie must be complete:
It must be able to reach the credits or end screen without requiring any further interaction
So, yes, you'll want to extend the input to clear that textbox and reach the credits. The published run does this as well. There are occasionally exceptions to this, but as a general rule this is always the safest option for a TAS. Even then, it's not a big deal at all if your submissions ever get it wrong. The input file can be modified and replaced very easily, and it'd never affect the judgement of the submission.
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
(2121)
Joined: 11/13/2006
Posts: 2793
Location: Northern California
I'll replace the file once I verify sync, but I will once again be waiting a number of days before re-accepting just to make sure that there are actually no more improvements to be had. I highly recommend double and possibly triple checking with the RTA community to see if there's anything left.
In the meantime... As I stated in my submission text, the ratings for this category have been steadily dropping over time. While a declining rating would normally be enough for a tier change, I can't help but think that these declining ratings are more due to the fatigue of so many Oddworld submissions as opposed to the run's actual entertainment value. That, combined with the general lack of thread feedback across nearly every previous iteration of this category, leads me to believe that we need to have a more objective discussion on entertainment.
How does everyone feel about the entertainment value of this run/game? Personally, I've felt a really sturdy Meh from the very first submission of this game. The major time cuts and route improvements due to the new glitches have made it a fair bit more interesting over time, but I'm still pretty firmly at Meh for this submission.
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
(2121)
Joined: 11/13/2006
Posts: 2793
Location: Northern California
For what it is, it's actually technically sound. It's just a shame that "what it is" is "completely unacceptable for publication due to using Game Genie codes". TASes like these are much better suited for the game's thread, where you can show them off without worry of being judged against the site rules, or alternately April Fool's Day, which is just a big opportunity for people to mess around with nonstandard runs, once again without any worry of being judged as a legit submission.
It's clear you have the ability to optimize and think outside the box, and I'd love to see that applied to an acceptable submission, though... Perhaps not SMB1. I feel like SMB1 is one of the only games, if not THE only game, on the site that has reached theoretical perfection. You'd have to discover something groundbreaking or do something absolutely incredible to get an SMB1 TAS published these days. I can't stop you from trying, of course, and I wouldn't want to do that anyway, but just keep in mind that if you intend to keep submitting, you're going to have a hard, if not impossible time if you stick to SMB1.
Whatever you do, though, stick around the community and keep at it!
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
(2121)
Joined: 11/13/2006
Posts: 2793
Location: Northern California
Think of it this way: Two teams that tie are almost certainly using the exact same routes on every stage. If you're having them aim for score as well, they'll have the exact same opportunities to gain score, meaning they'll end up with the same score as well, making the entire thing ultimately pointless.
If teams tie, let them tie.
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
(2121)
Joined: 11/13/2006
Posts: 2793
Location: Northern California
It's hard to judge a run's level of optimization without seeing it, but yeah, what you described is a great start. For a proper submission in the future, you'll just want to go as deep as possible doing that, making sure everything is as precise and as fast as you can possibly make it. We're not going to reject over mere frames of improvement or anything (we even accept runs with fairly large known improvements in specific cases), but the closer to perfect you get on a submission, the better.
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
(2121)
Joined: 11/13/2006
Posts: 2793
Location: Northern California
For the purposes of site submission...
PCSX2-rr is not supported by the site: You can't even submit movies made with it yet, as they wouldn't be recognized by the submission parser
Translation patches are not allowed even if it were able to be submitted. Cosmetic patches are not allowed either, though if the run syncs without them then they can be used for special encodes and such
Just aiming to beat an RTA run isn't enough, you have to optimize it as much as possible before submitting
Keep in mind that this is all only for a formal site submission. If you did it just to do it and host it on YouTube or some other video host, there's absolutely nothing wrong with it or any other TAS. You can talk about it or show it off here on the forums (there's a thread for KH2 here), but once you make a formal submission, it falls under the site rules and will be judged accordingly.
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
(2121)
Joined: 11/13/2006
Posts: 2793
Location: Northern California
That won't be necessary:
Xujhan wrote:
At the rate it's been going I should have an any% video to submit in about two weeks.
If there wasn't another any% coming for a much longer period of time, I would consider it, but with a new one coming very soon I don't see any reason to accept this as an improvement.
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
(2121)
Joined: 11/13/2006
Posts: 2793
Location: Northern California
TCRF wrote:
Enter the password "PQLLQ" ("pollo" is Spanish for "chicken") to enable the in-game stage select mode. During the game, press Start to pause it, then press Up or Down to change the level number. A level number will be displayed at the upper right corner of the screen. Unpause the game to jump to the selected level. Level 130 is the final entry and ends up with the credits roll.
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
(2121)
Joined: 11/13/2006
Posts: 2793
Location: Northern California
I'm giving this a Meh on entertainment. It kept my interest more than I was expecting, but around the midpoint I did kinda start skipping around a bit. I think my interest was more due to just the game being goofy in general. I love it when clearly alien monsters die with clearly human screams. Just inject that straight into my veins, thanks in advance.
Great looking run, definitely looks optimized for what it is, but it's still Ikari Warriors at the end of the day.
TASvideos Admin and acting Senior Judge 💙 | CohostwarmCabin wrote:
You shouldn't need a degree in computer science to get into this hobby.
