Experienced Forum User, Published Author, Site Admin, Skilled player
(1236)
Joined: 4/17/2010
Posts: 11269
Location: RU
Tremendous job gentlemen! Here's some cookies for you two.
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.
Experienced Forum User, Published Author, Site Admin, Skilled player
(1236)
Joined: 4/17/2010
Posts: 11269
Location: RU
Look at the startup part. None of you 2 is pushes, both are driving on a straight line. I'll be testing this track in your conditions.
The problem is, the movie rules say that a speed-oriented movie must beat all existing records. You don't seem to have checked this record.
Explanation of gear techniques from GameFAQs, for the reference (inluding the 4th gear tips):
Anyway, here's the legendary "Red Car Top Gear 4th Gear Trick" to shed extra seconds off your race time in the Cannibal (Red Car) This is not a glitch and is not a cheat- it is smart driving. In my opinion, this car was designed to run in 4th gear beyond what automatic transmission does for it. Automatic transmission for the Red Car is **bad**, we'll leave it at that. To be a true speed-master of the track, drive the red car in manual transmission.
Here's the trick-
Watch your speedometer, and let it run in 4th gear all the way up to around 215-225 Km/h (to figure out MPH for this equivalent, divide the Km/h by 1.6-- but I always drive in Km/h since it's a lot more accurate.)
Being in 4th gear up to around this speed will make your red car accelerate MUCH faster than if you had shifted earlier to 5th gear. It is also advised that you downshift to 3rd gear and then 4th for awhile if you hit an opposing car.
When going downhill, if your speed is below 220 Km/h, throw it into 4th gear. Memorize which sections of each tracks have decent downhill runs, and throw it into 4th gear when going down them for some extra speed. Your car should hit around 230 km/h or so, and once it gets around to that speed, shift to 5th gear to build further speed, as well as save
your fuel (running at higher revs in lower gears burns more fuel.)
When using a nitro, ALWAYS have the car in 5th gear, unless your speed is below 220 km/h from hitting a car or something. If that happens, shift down to 4th gear, and throw it back into 5th gear once you hit around 220-230 km/h, to make your nitro boost maximize its power, as well as conserving fuel.
In short, through using gears with the red car-
1st Gear- Shift up at the start of the race as soon as your tach fills to about halfway (audio cue also helps a good bit)
2nd Gear- Same as the first gear, shift up ASAP, you don't need to stay in this gear for long.
**3rd Gear- Should only be used if your speed is below 160 km/h. This will get you up to speed quicker. Shift up as soon as you hit 185-190-ish.
***4th Gear- The best gear to accelerate (gain speed) in. Hit it to 4th gear when you are above 190, and keep it in 4th until you hit around 215-225. Use this gear when going downhill if your speed is below 225. Once you get above 230 when going downhill, shift to 5th to get more speed and save fuel.
On long endurance races (pit stop races), you shouldn't run it in 4th gear long (try to use it only on downhill stretches-- like the end part of Tokyo, for example), so use 5th when around 215 or so to conserve your fuel.
On 3-4 lap "sprint" races, this really doesn't matter. Drive in 4th as much as you need to, sprint races are easy to win in the red car. You should not run out of fuel in any of them except Stonehenge, so drive as aggressively as you want. When using the red car, consider "sprint" races an easy 20 points.
***5th Gear- 5th gear should be used when your car hits around 215-225 km/h in speed, or, of course, when you are using a nitro. The 5th gear saves your car fuel and is the "holding speed" gear. If you're smart, you can actually hold speeds of 240-250 km/h in 5th gear for a long time.
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.
Experienced Forum User, Published Author, Site Admin, Skilled player
(1236)
Joined: 4/17/2010
Posts: 11269
Location: RU
Look like a good candidate. Just can you guys please rate it a bit harder?
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.
Experienced Forum User, Published Author, Site Admin, Skilled player
(1236)
Joined: 4/17/2010
Posts: 11269
Location: RU
You're losing several in-game seconds to the run published 14 years ago (on the Bordeaux track). How is it possible, given that run uses the same difficulty? It only uses one car, and starts from the first place, but from what I'm observing so far it's just plain faster.
[113] SNES Top Gear "one track" by Phil in 03:51.17
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.
Experienced Forum User, Published Author, Site Admin, Skilled player
(1236)
Joined: 4/17/2010
Posts: 11269
Location: RU
InfamousKnight wrote:
I get the feeling that tas tools going into upstream MAME is too good to be true..
I mean, being able to tas Windows xp would open a ton of possibilities..
Has any work been done for upstream MAME tas?
Not MAME, PCem! MAME doesn't emulate anything above 486. PCem has high chance of being added to bizhawk.
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.
These multiple account shenanigans are just petty.
Why?
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.
Experienced Forum User, Published Author, Site Admin, Skilled player
(1236)
Joined: 4/17/2010
Posts: 11269
Location: RU
<Mothrayas> "You've thought Mario was a cute little pet that makes everyone's life brighter?"
<Mothrayas> do they still keep people as pets in russia?
<feos> :)
<feos> we're all pets of the regime
<Masterjun> perfect execution feos
<Masterjun> that ":)" looks threatening like it rarely does
<Masterjun> it's all coming back to me
<Masterjun> help
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.
Experienced Forum User, Published Author, Site Admin, Skilled player
(1236)
Joined: 4/17/2010
Posts: 11269
Location: RU
It was tried and miserably failed. I wasn't really around, you'd need to ask the elders what exactly happened.
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.
Experienced Forum User, Published Author, Site Admin, Skilled player
(1236)
Joined: 4/17/2010
Posts: 11269
Location: RU
Oh boy. I finally watched it. This run is absolutely fucking brilliant.
You've thought after MrWint's bots this game's speedrunning was finallyofficiallyproperly dead? Meet 2 fucking framerules saved!
You've thought Mario was a cute little pet that makes everyone's life brighter? Meet this bloody massacre!
You've thought Wario was the evil twin of Mario? Meet this terror who is sick of his own game and wants to blast it into pieces!
You've thought you've seen all the tricks that don't waste time, and all the more elaborate ones that cost it? Meet this absurd mess that builds itself like a hallucination.
You've thought speedrunning was about going fast? Meet this run that flips the chessboard and plays its own game by slowing down, only to arrive exactly at the same time!
This is absolutely in every way a star to me, a NES TAS of the year, and I really want the warpless walkathon star to be moved over here.
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.
Experienced Forum User, Published Author, Site Admin, Skilled player
(1236)
Joined: 4/17/2010
Posts: 11269
Location: RU
I have a few questions. It absolutely doesn't sync on any other Dolphin version that has no av desync? Did av desync for the author too? Is it certain that av desync is not caused by missing audio samples, for which Ilari made this tool? Are savestates of other Dolphin versions incompatible?
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.
Experienced Forum User, Published Author, Site Admin, Skilled player
(1236)
Joined: 4/17/2010
Posts: 11269
Location: RU
Scepheo wrote:
There was a big discussion about two options. A very small group of people decided that there was to much discussion about those options, and a third one was enforced, without any discussion. The only reason there's no flame war in that thread is because everyone who disagreed with its content had given up on arguing.
Stop speaking for the entire community. Stop accusing staff who invests all their effort into handling everything, in being just a very small group of people. Stop exalting your subjective opinion over the entire community. Stop rewriting history.
No one decided that there was too much discussion. All the discussion that's been before my suggested solution has naturally stopped by itself. Literally nothing was enforced without discussion. You can pretend you're able to read everybody's mind and claim that they've given up. Yet you want the entire community to give up and accept your narrow-minded vision instead of the compromise that serves all the goals I listed.
Scepheo wrote:
And I'm not going to give you any examples of what you do label, because that is entirely unrelated to my suggestion.
Unfortunately it is exactly the main problem with your suggestion. You don't care what happens to everything and everyone after it's enforced and everyone who disagrees with you gives up.
Scepheo wrote:
I'm going to stop discussing this here though, as the only points you're actually addressing are the ones I'm not making.
This somewhat demonstrates how far you are ready to go when it comes to actually resolving the problem.
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.
Experienced Forum User, Published Author, Site Admin, Skilled player
(1236)
Joined: 4/17/2010
Posts: 11269
Location: RU
Sorry Aran. I tried reading a whole lot of your posts, and I simply can't. I keep giving up three lines into a sentence of 10.
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.
Experienced Forum User, Published Author, Site Admin, Skilled player
(1236)
Joined: 4/17/2010
Posts: 11269
Location: RU
Scepheo wrote:
I'm not the one using it as an argument: you are. You are ignoring half of TASVideos' audience. I'm just calling you out on it.
Still haven't.
Got it the first time. Still not actually responding to my points.
And again: I read it. You claim that thread somehow invalidates my earlier point that "clarity beats management difficulty". It doesn't: it has absolutely nothing to say about that point. If it did, you'd be able to quote the part where it did, instead of ignoring all my points.
That's the fourth time you're ignoring my points. I'm starting to think you don't actually have any arguments at all.
I guess I'll have to do it for you then.
When you have 2 camps with contradicting opinions on some subject, and each camp is approximately half of of your audience, blindly enforcing either opinion is impracticable, one-sided, unhelpful and narrow-minded. Especially when it's the audience who contributes and manages content in your community. You have to come up with solutions that would relax the tension, satisfy the overwhelming majority, be definable in a way that'd make sense to this majority, and serve on the long run. If you're unable to come up with such solutions, things will be getting worse, tension will keep increasing, flamewars will keep raising, and no one will be happy. One half will be annoyed with the enforced policy, another will be annoyed with the fact that they are disagreed with. And all this will be just a result of lazy staff who's unable to resolve the problem.
To resolve this problem, we had a staff conversation back in 2014. And we all agreed that whatever mess was introduced after retiring the "glitched" branch and removing the label from all "any%" branches, it should be fixed by using branch labels to describe the movies themselves, instead of describing the difference from "any%". The first thread you've seen contains the discussion where we try to point out all the problems we see and suggest solutions. The second thread suggests a solution. This solution was based on all the previous disagreement and agreement, and it was also commonly agreed upon, as lack of flamewar in that thread indicates.
The solution was to completely avoid enforcing either of the two originally suggested approaches. Because both of them are 100% based on subjective things called opinions. We stopped relying on opinions in labeling branches. We started relying on statistics. Whenever the approach of some branch is so common that it's default, we leave it without label. Because it's already implied. Whenever the approach of some branch is uncommon/rare/unique, we label it. When it's somewhere in-between, we label both variants.
But the important part of the original staff agreement was that "any%" is not something that needs to be addressed in the branch label. Because doing so results in the mess I'll show you after I find all the examples that convinced us back then. "Any%" is not descriptive. It tells nothing about gameplay conditions used to beat the game ASAP. Gameplay conditions are differences between branches. Difference between branches is what we need to mention if we want branch labels to help people orient in them.
So for internal game conditions we agreed to mention all options. And for external game conditions we agreed to mention the rare ones. Sorry that I don't have all the IRC logs from 2014. But it was all discussed in the usual #tasvideos room, and everyone could participate.
Now note that the main problem, "glitched vs. blank" and "blank vs. no X glitch", was resolved by not enforcing either of these options. We don't use "glitched" anymore for "major skip glitch" branches. And we don't use blank label anymore for any%. Because, as I said, these were subjective opinions of 2 halves of the community, and enforcing either of them is in no way a solution.
You may say that all this isn't contained in the second thread I linked. But all the main principles are, and I couldn't believe they remain unapprehended after having read that thread and remembering the events that led to it.
Scepheo wrote:
It's very simple, really: don't label goals that aren't actually goals of the movie.
All that means is that if a movie aims to do A, and it happens to do B, but only because doing B helps with A, you do not label the movie B.
And an example to round it off: if a Super Mario Bros. movie aims to do "fastest completion", and it happens to do "warps", but it only does "warps" because that helps with "fastest completion", you do not label the movie "warps".
And what do we label explicitly, and how? Use SMB3, Battletoads, Super Mario World, Super Metroid, and Super Mario 64 as examples of how perfectly your suggestion serves your plan.
Also note that "goals" is a rather broad term that you'll have to clearly define before demonstrating your suggested system in action, if you want the system to only care about this single aspect. The branching system since 2013 requires branch labels to not only accurately depict the goals and achievements, but to also differentiate each branch from others. Keep this requirement in mind when showing examples of your approach.
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.
Experienced Forum User, Published Author, Site Admin, Skilled player
(1236)
Joined: 4/17/2010
Posts: 11269
Location: RU
Scepheo wrote:
42%, to be exact. 44% don't. That's close enough to perfectly 50/50 to make the argument that "it is an issue" invalid.
Ah, so you're fine with ignoring half of our audience? OK. You've just been ignored as the unlucky half.
Scepheo wrote:
It is my opinion, which is shared by many of the other people that feel any% runs should be the default, unlabeled one. Yet another point where opinions are divided about 50/50.
You've clearly skipped the second thread I linked.
Scepheo wrote:
And again... It's the opinion of 50% of the people.
You've clearly skipped the second thread I linked.
Scepheo wrote:
I've seen that thread too. It has nothing to do with what I said.
Seeing is not enough. You will have to read it. Especially when you try to pretend it has nothing to do with this conversation.
Scepheo wrote:
You seem to have misunderstood the point of my post: you keep bringing up the same points. Almost all of them are subjective, and the exact thread/poll you've posted prove that opinions are divided 50/50. This renders those points useless to the discussion.
Okay. Now please go and read the second thread.
Scepheo wrote:
The only other point you bring up is that it would somehow be impossible to maintain the labeling system I (and many others) prefer. However, it's been done like that before, so it's clearly not. You've also failed to provide any evidence as to what has changed to make it impossible now.
That's a bold claim after having refused to actually read.
Scepheo wrote:
In short: I have yet to see a valid argument for the current system from your side.
Please point me to the part of that thread containing the evidence that "it ends up in unmanageable nightmare", because I can't find it.
I will. But I'll need you to post a full definition of the approach you defend, so I don't end up arguing with a false interpretation of it.
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.
Experienced Forum User, Published Author, Site Admin, Skilled player
(1236)
Joined: 4/17/2010
Posts: 11269
Location: RU
Scepheo wrote:
Many people (or at least I) don't consider this an issue.
This is known. As I said, unfortunately, 40% of the audience do consider this an issue.
Scepheo wrote:
In fact, this is a good thing: a any% run using warps should be obsoleted by a warpless run, if that somehow was faster.
Where are you getting this from? It's not what the Judge Guidelines say. They say that double obsoletion (for example any% obsoleting any% and 100% at once) can happen when the new run achieves the goals of both other branches. Cross-branch obsoletion is another thing, and it only happens when the branches are similar enough.
Scepheo wrote:
Any% is "as fast as possible". Any other goals it overlaps with are incidental and don't matter.
Thanks for your binary opinion. Are you sure that if it doesn't matter for you, all other people should be forced to stop caring?
Scepheo wrote:
This is the only valid reason I've seen given for the current labeling system. There's two problems with it though:
1. I've yet to see any proof of this.
I guess you've missed this thread then. Which is strange, because you've posted there.
Scepheo wrote:
I don't feel you've ever actually explained it, though.
Thanks for sharing your feelings.
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.
Experienced Forum User, Published Author, Site Admin, Skilled player
(1236)
Joined: 4/17/2010
Posts: 11269
Location: RU
If hardest difficulty adds variety and complexity, is harder to play and to TAS, use that. If it only makes things tedious and long, use easiest. If unsure, use hardest.
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.
Experienced Forum User, Published Author, Site Admin, Skilled player
(1236)
Joined: 4/17/2010
Posts: 11269
Location: RU
HappyLee wrote:
Sorry but I just saw this topic and the branch name changes of my familiar movies. I have to say honestly I prefer the old way of naming branches, and to me, branch names should be as short, simple and elegant as possible.
How old? Your warped run was renamed 4 years ago. You mean even older? Community decided that the older approach doesn't work. I was okay with the older one. But we had to invent something new, and it was invented. Note that these threads mostly talk about "glitched". But the problem is identical. What happens when "glitched" branch is shorter than "warps"? We change the warped branch from blank to "warps", and remove the branch from "glitched", since it's now faster? 40% of our members insisted on that. Other 40% insisted on keeping the warped branch labelless, and labeling the "glitched" one. Really, such problems can't be resolved in a way that's perfect to each contradicting camps. So yes, we invented a compromise.
But as was pointed out by Nach in this thread, even then we'd have to rename branches constantly, every time a new branch appears. So we decided to give branches names that'd stick for years.
HappyLee wrote:
Nach said "you know what each run is about by looking at it's title", but we also have tags which also contain information and are helpful to newcomers.
I think branch should reflect goals in the simplest form, and we don't need an artificial branch for a simple "any%" goal.
It's impossible to reflect them in the simplest form when you have 2 camps with contradicting labeling preferences, tons of variations of branches, up to 5 branches per game, new branches appearing every year, and if the site grows, even more branches per game potentially.
HappyLee wrote:
For example:
NES Super Mario Bros. (JPN/USA PRG0) "warps" in 04:57.31 by HappyLee
I don't think this "warps" branch is necessary, because the goal is equivalent to "any%", and "warps" is just one of the basic things Mario can do, not a limitation or special goal choice. We don't used to have it and it worked just fine.
It's only equivalent to any% until a major skip glitch is found. And only for this game. For other games, any% is equivalent to other goals. As mentioned in the first post of this thread, if we always drop the label of any%, and label all other branches, it ends up in unmanageable nightmare.
HappyLee wrote:
For example, we have:
NES Contra (JPN) "pacifist" by Soig & zyr2288 in 09:29.08
and
NES Contra (JPN) by zyr2288 in 08:51.73
These two branch names are simple and elegant. We don't have to call the latter:
NES Contra (JPN) "kills" by zyr2288 in 08:51.73
Just as it would be a disaster to call the current SMB run:
NES Super Mario Bros. (JPN/USA PRG0) "warps, jumps, runs, kills, no-powerup, not maximum coins" in 04:57.31 by HappyLee
But why "kills" is not a branch? Because it uses a mechanic that's so common that we consider it default. Similarly, "jumps, runs" are standard mechanics in speedruns. We already imply them. But for any mechanic that can become a new branch, we add a label.
Not so long ago Memory asked me, why don't we consider 1p and 2p internal goals in SMB and why don't we label them. Here's the answer I came up with:
These goals are highlighted, because they introduce gameplay differences into TASes that are significant enough to be published as separate branches.
"Warps" is not a default goal, because "warpless" is also really common, and not every game has warps. Same with "1 player" and "2 players". We already discussed this in details in this thread.
HappyLee wrote:
However, "warpless walkathon" is totally necessary, because with "warpless" removed from the goal, it's just "walkathon", and walkathon's a lot faster with warps.
I explained above why we can't leave all the fastest branches without labels.
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.
Experienced Forum User, Published Author, Site Admin, Skilled player
(1236)
Joined: 4/17/2010
Posts: 11269
Location: RU
Aran Jaeger wrote:
You two, can we please stop this now and get back on the topic ''What is the point in submission polling?'', unless I'm misunderstanding and this is related to it? Or we could also split the thread up, which I think would also be an option.
I already suggested that on IRC, a huge part of this thread clearly belongs to the "Arguments regarding arguing" topic. Too bad it's locked.
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.
Experienced Forum User, Published Author, Site Admin, Skilled player
(1236)
Joined: 4/17/2010
Posts: 11269
Location: RU
Nach wrote:
Statistics don't work for explaining anything at all
p4wn3r wrote:
If statistics does not matter at all, there's no problem if the data is fake in the first place.
Uh, really? Can you please not randomly substitute words?
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.
Experienced Forum User, Published Author, Site Admin, Skilled player
(1236)
Joined: 4/17/2010
Posts: 11269
Location: RU
ais523 wrote:
This is down to search algorithms. There are some search orders you can use that are guaranteed to find the shortest output first; the simplest is to try all the possible inputs in length order (except for ones which have a prefix that turned out to be slower than another way of doing the same thing). I think there may be games where this is simple enough that it's actually within the capabilities of a powerful modern computer.
I have some more complex algorithms in mind that would have the same properties but be more efficient. The problem is just finding time to getting around to implementing them.
Yes, I understand that state that's been achieved previously in longer time can be disregarded, along with its input, if shorter input reaching it is found. And indeed I see that it helps to avoid truly exhaustive search. I just want to understand problems that raise after this method is applied, and how hard/easy they can be solved. Can you be more specific on what such an algo would have to do?
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.
Experienced Forum User, Published Author, Site Admin, Skilled player
(1236)
Joined: 4/17/2010
Posts: 11269
Location: RU
DrD2k9 wrote:
Is there enough perceived confusion about polls (specifically, but not limited to the 'meh' option) for the staff to consider a change of any type? Or is the general consensus that what we have is good enough and doesn't need changing?
Personally, I think the available answers unfortunately cause the poll to be utilized by voters from both a perspective of entertainment (as it's written/intended) but also as meaning "should this be published?" (which is not up to the community, but the judge).
Problems with submission poll do exist. I quoted others talking about these problems in the first half of this post: Post #421751. These are the problems that make the current system imperfect. Less reliable than we'd want.
Why wasn't the poll question changed back then? There wasn't staff agreement about how serious the problems with it are, how to solve them best, whether the tier system needs changes, etc. This minor aspect, poll question, has fundamental site vision backing it up. That vision was introduced in 2012 and opened some possibilities for boring movies and for additional branches. Its main idea is promoting entertaining content.
If there's not enough entertainment, your run goes to Vault if it fits into the category limit. If there's enough entertainment, your arbitrary branches can be published. For any% and 100% that are entertaining, this is just a way to reach out to the viewer easier and to get more feedback as a result.
Untying the poll from entertainment criterion won't help judges in any way. Entertainment value of the movie still has to be assessed. With quite a few exceptions, 80%+ support is most likely Moons. But:
- Should this boring movie be published?
- Yes! To Vault!
You're not getting useful feedback that helps with the judging if the question is changed to that.
I know you didn't suggest it to change like this. But this is still stuck in people's minds very hard.
And that is exactly the problem! Human factor. People with spoil their "objectively reasonable" votes with all sorts of irrelevant things and still find them fitting. This is the reality that can't be fixed or worked around. We are asking them subjective questions, they provide subjective answers. It is okay to receive irrational feedback, because rational stuff we can measure without any feedback at all! But this site is viewer oriented, so irrationally bad and irrationally good impression is something we need aside from plain record keeping. See this thread for reasons:
Thread #20113: Origins of TASVideos' name - split 14293, 20112DrD2k9 wrote:
I think (as I suggested before) that a poll regarding the degree of entertainment may be better than the yes/no/meh options for "did you find this entertaining?"
As explained above, entertainment is irrational, and we want this irrational feedback in addition to rational feedback about optimality. Splitting 3 irrational options to 5 won't change anything. Renaming them won't change anything. Removing them will probably also fail to improve anything. Making the answers public clearly makes things worse, because anyone will be able to flash-mob pro or against someone's vote, so people will just start avoiding voting altogether.
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.
Experienced Forum User, Published Author, Site Admin, Skilled player
(1236)
Joined: 4/17/2010
Posts: 11269
Location: RU
Sorry for sweetening up, but this is exactly the thing I tried to address in the post about professionalism. The world is imperfect, it always was, it will forever be. People are imperfect. It is impossible to fix or work around all problems. It is impossible to prevent all abuse. It is impossible to make everyone happy.
Getting upset over getting imperfect outcome despite doing everything perfectly is natural. And at the same time it's a psychological problem. While to some degree it can be helped by wise actions made by others, expecting everyone to be wise is unrealistic.
Remaining calm when your perfect peace of work is being unreasonably criticized may feel like a victim being blamed. And neither of these helps anyone with anything. Unwise decisions coming out of this feeling won't help anyone with anything.
This is objective reality, it can't be directly fought. But it opens possibilities for raising upon yourself. Even if this doesn't improve the situation, it improves the notion you have.
General "you", no one in particular.
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.
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.
Experienced Forum User, Published Author, Site Admin, Skilled player
(1236)
Joined: 4/17/2010
Posts: 11269
Location: RU
Jabo video plugin for N64 TASes has been deprecated.Thread #20237: Deprecation of Jabo video plugin
All TAS submissions using Jabo from now on will be encoded with minimal and default settings.
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.
Experienced Forum User, Published Author, Site Admin, Skilled player
(1236)
Joined: 4/17/2010
Posts: 11269
Location: RU
Jabo video plugin for N64 TASes has been deprecated.Thread #20237: Deprecation of Jabo video plugin
All TAS submissions using Jabo from now on will be encoded with minimal and default settings.
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.
