page  1234567891011121314 -> 1 ... 14 ->
--
--
List results:
Search options:
Use \ before commas in usernames
Edit history:
LLCoolDave: 2009-06-14 06:59:05 am
The intent of this topic is to publically discuss some of the sketchier parts of the SDA rules and suggest changes and improvements. The overall goal is to streamline the rules, get rid of unwarranted arbitrary restrictions and exceptions and rework legacy rules that are not on par with the current SDA ideology. Although this discussion is officially sanctioned (and I expect this topic to be stickied if there is enough participation) be aware that there is no guarantee that any of the changes agreed upon in here will become part of the rules, nor is there any guarantee that every opinion can be considered fairly and equally. In fact, some of the points to be discussed have been disagreed on for years now and we don't expect to find a solution that will satisfy all participants or even the majority of them.

Take all opinion and comments voiced in this thread to be a very personal stance unless noted otherwise, uncluding posts by staff members. I hope for this discussion to be on topic and as neutral as possible. You are free to disagree with any part of the discussion, but try to refrain from any form of personal attacks. Expect there to be more and harsher moderation than in other parts of the Forum to ensure a productive discussion. We are willing to listen to all of your concerns as long as you stay polite.

List of Topics to discuss:

1.) Treatment of currently published or in-submission runs affected by the proposed rule changes.
  Issue: Several runs may be breaking the rules in their proposed new form, several runs may have their timing or categorization affected by the new rules in a significant way. This may interact with previously obsoleted runs as well.
  Goal: Finding guidelines for treating affected runs fairly without counteracting the goal of reworking the rules. Finding guidelines for reinstating previously obsoleted runs should the current published run become illegal under the new rules.

2.) Reworking rules and faq text according to the new proposed rules, rewording parts of the rules and faq that are or have in the past been considered to be inaccurate or hard to understand.

3.) Reworking the save penalty for manual saves in games that allow you to save anywhere
  Issue: People regularly are confused about both when and why the penalty is applied. Ever since the introduction of the rule there have been people voting for getting rid of it as well as people who think the penalty is too small.
  Goal: Discussing the necessaity as well as the benefits and downsides of such a rule. Discussing the saves it should be applied to as well as the size of the penalty. Finding alternative rules that may better achieve the goals of this rule.

4.) Clearly specifying the scripts that are to be allowed.
  Issue: Several games still have a grandfather clause of being allowed to use Autohotkey-Scripts that are now officially banned. A couple of runs have used scripts that several people consider to be against the spirit of the site.
  Goal: Reaching a consense in how much scripting is to be allowed in runs. Finding a way to distinguish legal and illegal scripting that leaves a low amount of borderline cases.

5.) Dealing with game specific rulings
  Issue: Several games have gotten special rules and exceptions for various reasons in the past. Although, due to the variety of games and runs on the site, it is not possible to cover all cases with a set of basic rules, several games have special rules that appear to be unwarranted under the current ideology.
  Goal: Getting rid of as many unwarranted game specific rules as possible to integrate them into the basic rules and avoiding arbitrary individual rulings.
    a.) Half-Life
    b.) Half-Life 2
    c.) Portal
    d.) If you find any other game with specific ruling that you feel should be removed, post it and I'll add it to this list. This list in no way claims to be comprehensive.

6.) Dealing with very similar categories
  Issue: For some games there may be clearly defined seperate categories that are however very similar. For instance in Banjo-Kazooie the any% run requires you to pick up almost as much as a 100% run, and the difference is fairly small for a comparably long game. Similaarly, some games allow for character choice which has a pretty small influence on gameplay. Does it make sense to accept these categories seperately in this case? However, wouldn't this also lead to more semi-arbitrary game specific rulings?

7.) In-Game Timer vs Real Time
  Issue: There are people who feel that every run should be timed manually. In-game timers are usually what dedicated communities use to compare their times, which helps making SDA runs comparable to runs done outside SDA. It is currently somewhat well understood what an acurate timer is, but not clearly defined in the rules.
  Goal: Find a better definition of acurate timers (in SDA eyes) to avoid misconceptions. Discussing the benefits of using one timing over the other in specific situations.

If you have any other item of discussion that you feel should be on this list, please suggest adding it. As there are a lot of topics to discuss in a single thread, please Always clearly mark what topic you are discussing. This will be a great help in keeping this thread focused.

Last Updated: Jun 6th 13:58 GMT
Thread title:  
Post reserved.
Post reserved.
Post reserved for proposed updated rules text.
Post reserved for proposed updated faq text.
Edit history:
groobo: 2009-06-14 04:47:12 am
boss
3) Finding alternative rules that may better achieve the goals of this rule? Why? Why even modify this rule? Get rid of it ASAP. In theory the point of this was to prevent people doing runs in games with quick saving from going apeshit. Problem is that even if this rule would never exist, nobody would. I know someone will mention half-life in a minute, so let me ask this: what is the real goal of this site? No matter if you up the penalty to ten seconds or remove it completely, the runs will look exactly the same.

4) Although I was against this, my stance on the subject changed over time - allow scripts for ALL games. Also allow turbo controllers (for all the haters who can play mega man without scripts so why allow them in other games) - I have this stupid little grin on my face right now, so I'm not entirely sure if I'm not joking. It's not like you absolutely have to use them anyway. If you really have to, make them into a separate category - even death abuse got one for the most bullshit reason ever - and I still curse Mike for it. /hate

But no, really, allow scripts for all games.

5) Subject number four when solved will give us an answer for this one. /spoiler (the staff agreed only on all suggestions thus far and this should be only discussed if #4 gets the thumbs down, which btw would be a real shocker)
1) The fairest way is just to keep runs that break the rules on the site, with a note explaining the rules that applied to them at the time on the game page, but not accept improvements using those rules. Naturally, if someone can improve on an old run with a new, legal one, then the old one should be taken off the site and we should rejoice.
2) I've never found anything in the FAQ and rules text hard to understand, though some parts are intentionally vague to allow admin discretion.
3)
Quote:
No matter if you up the penalty to ten seconds or remove it completely, the runs will look exactly the same.
I totally disagree, Groobo. A ten second penalty would mean doing nearly every level of a typical FPS as its own segment, only manual saving to allow the execution of extremely difficult tricks, which would make play notably less perfect. Personally if we're going to impose a penalty I'd favour something in the range of 2-4 seconds. And if there was no penalty, I would go apeshit with quicksaves; I mean, why would you accept imperfections in your run if there's no disincentive to segment further to remove them?
4) Hard one, this. I think I'm against allowing scripts altogether, but then I don't really understand what they are capable of since I've never used them. Theoretically, couldn't you effectively TAS a game with scripts my creating a massive mega-script that when executed runs the whole level automatically? Or am I totally overestimating what it's possible to do?
5) This is just a mixture of 4 and 1, really. Leave the grandfather rules, but hope somebody will beat the runs with the new rules allowing the grandfathered ones to be put to death forever.
Invisible avatar
1) If old runs are breaking the new rules, just leave them laying around with a clause about it. For changed timing, well, some people might go crazy. And to those people I ask, what's the difference if your run gets even 20 seconds added to it? The quality of the run is still the same. Just because the quantification of the quality (the time) changes doesn't change the quality of the run.

2) FAQ and rules are pretty understandable.

3) Get it up to at least a second. I'd vote for 2, frankly. And to reply to groobo, larger save penalty will achieve two goals: stupid people will stop calling multi-segment runs devoid of skill; and it will combat the 'save the moment something goes sour'. The reason the saving right now is so similar to what it would be without the penalty is precisely *because* it's a penalty too low. Finally, I'd love a little more challenge in my segmented runs myself - I know I'd think twice about using some points for saving in DX if the penalty was 2 seconds 8). Heck, I know I would add several more saves in my current Deus Ex run if there was no penalty (God, that game is random).

4) No movement scripts. I see no harm in using weapon change scripts, and those are mostly transparent to the watcher and hence it's impossible to enforce the ban of them anyway.

Quote:
Theoretically, couldn't you effectively TAS a game with scripts my creating a massive mega-script that when executed runs the whole level automatically? Or am I totally overestimating what it's possible to do?


Theoretically, though that would be outside of anyone's scope - simply too much hassle. The theoretical possibility depends on the game much, too.
So here's my opinion on these first issues:

1.) I feel the best solution is to keep around the now illegal runs but discontinue the relevant categories. Should there be obsoleted runs that still fit the play quality of being accepted to SDA they should be reinstated as the "current record", for whatever it is worth. Keep in mind that SDA is not a place to track records, so this is not actually a serious issues. Although several runs on SDA may qualify as world records, we never have and never will claim to host record runs. Timing runs is meant to be consistent, not "correct", so it doesn't really matter what standards are used as basis for what the currently best run on the site is, it is however advisable for having these standards be as consistant as reasonable accross all games. I don't see a problem of reposting a previously obsoleted run if it passes verification again.

3.) Something that many people seem to completely miss about the save penalty is that it doesn't (only) exist to be against super massive segmenting. Many PC Games, especially when using some sort of quicksave, process some input before the first visible frame is displayed after loading a save, allowing for some movement during a time of the run that is both untimed and immensly hard to time (how do you determine just how much of the movement on loading is happening in the non visible part?). If you stick the visible parts of segments right after each other in games where this applies, you will see minor warps. They are the only, even if hard to notice, way to tell at least some of the segmenting going on in the half-life half hour run, but are even more obvious in games such as Jedi Outcast. The save penalty was (at least partly) introduced to offset the gain by this, and I am fairly certain it was originally my idea. In any case, the use of it to discourage heavy segmenting is clearly up for debate, and I personally wouldn't mind seeing an increase or some other, maybe not timing related, rule to achieve that. However, the timing issue still needs to at least be adressed. If we just completely remove this rule one could on some of the worse offender games move during the loading and immediately resave with no penalty, getting 0.2 seconds or so of movement for a single frame of timing. I think this is pretty strange, to say the least. Another issue with the rule as it stands now is that autosaves are not taken into account, although they are on many games nothing more than an automated quicksave with the same quirks when loaded.
The reason this rule was designed for PC Games originally is just that console games usually don't show any of this behavior due to the way most of them work.

4.) I am entirely against automation and scripts, but the main problem is defining what counts as a script and what doesn't. A popular quake "script" is "impulse 4; wait; impulse 5;" which first selects tha standard nailgun, if available, then selects the super nailgun, if available. As there is usually no reason to use the nailgun if you have the super nailgun, this script allows to bind both nailguns to the same button, freeing another key to be used for other actions. I'm not entirely sure if this should already count as a script or not, but thee 180° turning script used with the tau gun in the half-life run is really upsetting me. We definitely do not want to remove the ability to customly map controls, but I can't really say where the, well, cutoff should be made, because not all controls that are unproblematic are always configurable with a menu, and not all menu configurable controls are always single action commands.

5.) My general stance is to get rid of these special treatments under application of 1.) It makes no sense to allow Half-Life to use AHK-Scripts and other games are forbidden, and I don't think we will see a return of AHK anytime soon, though the decision is obviously not up to me.

Wish I could say more, but I don't have too much time today.
boss
I'll ask again: what's the goal of the site? A show of skill, or entertainment? TBH it's balanced right now and what you're trying to do is to hurt one side in the other's desperate favour. Remember that subjects 3,4, and 5 affect only a handful of games (all affect the same games in fact). Enhasa is always saying how his goal is to get as many runs up on the site as possible which, if any of your suggestions passes, will be one big fucking paradox.

Leave the grandfather clause and never return to this subject again, please.
Quote from dex:
The quality of the run is still the same. Just because the quantification of the quality (the time) changes doesn't change the quality of the run.


I assume your talking about changing the save penalty here? Well then it does change the quality of the run, massively, because it affects whether the decisions about where to segment make sense or not.

Quote from LLCoolDave:
4.) I am entirely against automation and scripts, but the main problem is defining what counts as a script and what doesn't. A popular quake "script" is "impulse 4; wait; impulse 5;" which first selects tha standard nailgun, if available, then selects the super nailgun, if available. As there is usually no reason to use the nailgun if you have the super nailgun, this script allows to bind both nailguns to the same button, freeing another key to be used for other actions. I'm not entirely sure if this should already count as a script or not, but thee 180° turning script used with the tau gun in the half-life run is really upsetting me. We definitely do not want to remove the ability to customly map controls, but I can't really say where the, well, cutoff should be made, because not all controls that are unproblematic are always configurable with a menu, and not all menu configurable controls are always single action commands.


Concerns about TASing notwithstanding, this sums up my view on the matter perfectly.

Issue 6) I would not have seperate categories, but allow people to do 100%/low%/different difficulty/different character etc. runs and have the verifiers determine which run is better when there are two of different categories competing. If the differences are genuinely small, this shouldn't be hard to determine.

Issue 7) I think manual timing and in-game timing should be seperate categories with the caveat that, as with my reply above, if manual-timed runs are directly comparable to in-game timed runs, verifiers should determine which run is actually better rather than just taking the one with the lower time. I personally dislike the use of in-game timing for games with lots of menu time where planning to reduce the menu time would significantly affect the run, and was unhappy with being forced by Mike to use the in-game timer for my Hitman: Blood Money SS run when in my opinion optimising weapon choices and balancing the time saved by weapon upgrades against the menu time spent buying them would've been the most interesting part of an otherwise fairly bland run - which as a result of in-game timing, is now broken up by lots of slow buying of everything available at every level transition. I know there have been other games in the past with similar issues that have caused some serious contention, but they're not games I know anything about so I'll leave it to others to comment on them.

The point has been raised that more manual timing means more work for Mike. A suggested solution is to have a Time Team responsible for timing (and indeed to encourage verifiers to do it and trust them to get it right) to reduce the work Mike has to do. I approve of this suggestion.

Edit: Groobo, why the needless hostility? I entirely fail to see how Dave's suggestions sacrifice entertainment for skill, which is what I presume you're saying. Stuff like 180 degree turn scripts look retarded and non-human, and so hurt entertainment - skill doesn't even come into that debate. As for the save penalty, pretty much any game with saving anywhere has warping upon saving and loading, and IMO one of the biggest reasons to avoid insane amounts of segmentation is that warping looks shit - again an entertainment issue that has nothing to do with skill. And besides, I think less segmentation = greater display of skill = more entertainment, I don't see one being sacrificed for the other. Where do you think entertainment is lost?  Huh?

And, uh, how do ANY of Dave's suggestions have any impact at all on how many runs get up on the site?
Invisible avatar
Quote from ExplodingCabbage:
Quote from dex:
The quality of the run is still the same. Just because the quantification of the quality (the time) changes doesn't change the quality of the run.
I assume your talking about changing the save penalty here? Well then it does change the quality of the run, massively, because it affects whether the decisions about where to segment make sense or not.

I was talking more in the quality of gameplay sense :P. Also, rethinking the issue, I think that old runs shouldn't need to be re-timed under new rules: it would just put more hassle on Mike's shoulders, and is pretty much unnecessary.
Quote from ExplodingCabbage:
Edit: Groobo, why the needless hostility? I entirely fail to see how Dave's suggestions sacrifice entertainment for skill, which is what I presume you're saying. Stuff like 180 degree turn scripts look retarded and non-human, and so hurt entertainment - skill doesn't even come into that debate. As for the save penalty, pretty much any game with saving anywhere has warping upon saving and loading, and IMO one of the biggest reasons to avoid insane amounts of segmentation is that warping looks shit - again an entertainment issue that has nothing to do with skill. And besides, I think less segmentation = greater display of skill = more entertainment, I don't see one being sacrificed for the other. Where do you think entertainment is lost?  Huh?

And, uh, how do ANY of Dave's suggestions have any impact at all on how many runs get up on the site?

This. Also, no need to start cursing around.
regardless of what the rules say it's really just a matter of asking mike on a case by case basis.  this is what makes spelling out rules kind of a waste of energy imo

Cheesy
TIOLET!
On issue no. 3.

I really think that the current save penalty of 0.5 seconds is effectively useless and should be removed. You can see a trend that segmented runs are done by using an increasing amount of segments, so the penalty obviously fails at discouraging runners from using a large number of segments. If the penalty is raised I think it becomes counterproductive in that some tricks or shortcuts might be left out from a run simply because the penalty adds too much time to the run, and in turn forces the runner to leave things out that requires too much luck to pull off with fewer segments.

That last sentence was a train wreck but I hope you understand what I tried to say.
Quote from xXIkuto:
regardless of what the rules say it's really just a matter of asking mike on a case by case basis.  this is what makes spelling out rules kind of a waste of energy imo


Agreed that there's no need (and indeed no reason) to have extremely precise, rigorous rules, but most of the debates here are sufficiently general that they're worth having. When dealing with specific cases, Mike operates like a court judge, determining how to interpret points of law and using common sense when no law applies - but this is a debate about what the laws should be in the first place.

Re: Kibbo's trainwreck, I don't think a 2, 3 or even 4 second penalty would really prevent many tricks. How many tricks are there that only save 4 seconds or less, and how many of them are so difficult they can only be done at the start of a segment? Not many, I expect.
TIOLET!
Quote:
How many tricks are there that only save 4 seconds or less, and how many of them are so difficult they can only be done at the start of a segment? Not many, I expect.


This we can't know for sure, but it can definitely happen in a lot of games.
My feelings on The Demon Rush
First off, yes, I let LLCoolDave make this topic because Enhasa and I were planning on making a revision to the rules and FAQ pages for things like real name being optional now.

I should probably answer more of these, but for now I'll just answer a couple.

Scripts (well, movement scripts): Enhasa and I discussed this before, and we agreed that disallowing turbo fire while allowing scripts was contradictory because they're both tools that let you do things that are humanly impossible (don't tell me that turning 180 degrees perfectly every time is possible). We just left scripts because it would be too much of a hassle making a ruling that took out scripts without taking runs down or pissing people off even though neither one of us agrees with scripts. So yes, groobo, we already have a paradox.

6. This has already been addressed. http://speeddemosarchive.com/ConkersBadFurDay.html It's a case-by-case basis, but basically if two categories are already very similar, we just go with one.

7. This issue isn't the most pressing matter because in-game timer is fine for most of the games submitted so far. I have no problem with making real-time category a separate category for some games, but it would definitely have to be on a case-by-case basis (i.e. Mega Man 9) because most games don't warrant separate categories for timing.
@_@
Most of these issues I'm kind of unopinionated on and can work with whatever the majority decides, but I guess I can state my thoughts on a couple.

3.) I'm one of those people who are for the save penalty being removed. I think the way it is currently is both ineffective and unnecessary. You can save 100 times and not even add a minute onto the final time, and if the run's length is so long it really requires that much segmentation, then either the 50 seconds added won't even be seen (seconds are dropped off over 3 hours) or there's much more time saved than is penalized anyway. Increasing the penalty will dilute its original purpose, which is stated to have been to keep people from using excessive segmenting, by imposing harsher penalties on those runs which are NOT overly segmented. If you don't raise the penalty enough to sufficiently penalize over-segmented runs, then there's no point in raising it at all. The whole thing really serves no purpose. It needs to be removed. In my opinion, a run with a needlessly high number of segments and low runtime punishes itself. A lot of people will look at that and think "Oh wow! They got x:xx:xx on xxxxxx game! Oh wait... Over 100 segments? No thanks." If each segment is that short, the entertainment value is possibly that much lower and the skill displayed could indirectly be that much lower also, at least in the eyes of a casual viewer (and maybe even some runners that have done long single-segment runs).

Quote from ExplodingCabbage:
I don't think a 2, 3 or even 4 second penalty would really prevent many tricks. How many tricks are there that only save 4 seconds or less, and how many of them are so difficult they can only be done at the start of a segment? Not many, I expect.


This takes looking at it more than one segment at a time imo. If you segment one part in order to pull off a 20-second timesaver, with 4 more parts that were simply segmented for better optimization your timesaver is already nullified with a 4 second penalty on EACH segment.



7.) I personally am under the belief that the more in-game timer usage we can allow, the better. I believe that SDA would get MORE complaints if we used real timing for everything, especially in games where an in-game timer is distinctly displayed upon the end of the game and the SDA timing turns out to be different. If anything, more random people would make a guest post on the forums saying "There's a typo in the runtime" no matter how many places we said "WE TIME USING REAL TIME NOT IN-GAME TIME!!!!"

The time of a run serves multiple purposes for me. Not only is it obviously the way to gauge how fast the run is, but it's also to compare different runs of the same game. For BOTH purposes, using an in-game time is just as possible (and in many cases, easier) as using real time. If you complete a game and it used its own abstract method of timekeeping, say the final screen said "Completion time: 350 units." I would even be fine with the time displayed on the webpage as "350 units" with an approximation of the time in parentheses beside it somewhere instead of just "x:xx:xx" since the units time is an easy way to compare. If I completed the game at home in 725 units, I would look at the run and be just as impressed as if I used real time.

But since the real goal listed was primarily "Find a better definition of acurate timers (in SDA eyes) to avoid misconceptions," I probably digress. I think the rules pages spell it out enough for me to understand what an accurate timer should be. If it cuts off the seconds, don't use it. If it stops during time where SDA would normally time, most notably dialogues and inventory screens, don't use it. I guess MM9 would break the 2nd rule though, so for that to stand the description does need a rework (or as Mike said, separate categories which solves this case). Trying to officially describe something like that in a rules page is difficult though, especially when we seem to rely mostly on common sense when judging those types of things, as well as whether to use in-game or real time in the first place.
gamelogs.org
–don't see a reason to keep the segment penalty. it doesn't do much.

–use ingame timers whenever they're accurate. the less individual rulings for games there are, the better.

–no scripts of any kind should be allowed. seems absolutely ridiculous to even discuss them. this isn't tasvideos.

–grandfather clauses should of course be used whenever old runs break current rules. said old runs should be left up until they're beaten.
Waiting hurts my soul...
I'll post my opinions on each later, but I thought now would be a good time to bring up an 8th topic.  Death in games as a separate category.  This is based on an old philosphy that deaths is speedruns shouldn't happen.  Then came death warping, which was penalized and noted as such.  Later the penalty was done away with and death abuse became a separate category.  This was simplified recently to just noting the game had deaths, but it still was a separate category.  I may be wrong on that timeline, but that's the general idea of where this all started.

Since it's based on such an old rule, which has changed slightly over the years, I think it needs to be reevaluated wholely.

My suggestion would be to remove the separate category altogether and just let the fastest run stand.  Deaths in a run  should be something the verifiers decide if it's acceptable or not.  I know in some cases deaths can skip large portions in a game, in which case I suggest it be considered a skip glitch, or maybe make a run without the skip death be a 100% or the skip death run low%?  In those cases though, game by game is about the only way to judge, but that's the kind of thing we're discussing here right?
Edit history:
groobo: 2009-06-14 12:16:07 pm
boss
Quote from ExplodingCabbage:
Stuff like 180 degree turn scripts look retarded and non-human, and so hurt entertainment

This is your opinion, not the ultimate truth. I for one very much enjoyed it and there are more people who did.

Quote:
As for the save penalty, pretty much any game with saving anywhere has warping upon saving and loading

When it's really obvious, then it gets a penalty.

Quote:
IMO one of the biggest reasons to avoid insane amounts of segmentation is that warping looks shit

Actually that's the exact same reason why nobody would go apeshit with segmenting. I mean I did runs, you know. If not for the rule all my segmented pc runs would've had two, maybe three segments more - there's no point in splitting parts into more segments than necessary. All this rule does is up the time shown on the game page and in the statid and who cares about that, really. I will go even further: eg. because of this rule I had to scrap out some small time saving tricks in some of my runs and that just contradicts my whole idea of speedrunning.

Quote:
And besides, I think less segmentation = greater display of skill = more entertainment

Again, just your opinion. Doesn't really matter if it's ss or segmented as long as it's entertaining. Having a debate over which one is MORE entertaining is stupid (dex won't let me curse here  :(, although it's highly appropriate in this case).

Quote from mikwuyma:
Scripts (well, movement scripts): Enhasa and I discussed this before, and we agreed that disallowing turbo fire while allowing scripts was contradictory because they're both tools that let you do things that are humanly impossible (don't tell me that turning 180 degrees perfectly every time is possible). We just left scripts because it would be too much of a hassle making a ruling that took out scripts without taking runs down or pissing people off even though neither one of us agrees with scripts. So yes, groobo, we already have a paradox.

Well, assuming the WE WANT MORE RUNS UP ON THE SITE statement is still valid, this perfectly makes sense. (regarding 3&4) If we'd bring even more restrictions we'd potentially prevent a new run(s) from getting on the site, also probably we'd successfully discourage some runners from contributing. Either leave the rules as they are right now or let them go to hell. It won't do any harm, but going the other way probably will.
Quote from NoiseCrash:
3.) I'm one of those people who are for the save penalty being removed. I think the way it is currently is both ineffective and unnecessary.


Disagree. If it was taken away, I'd never do a segmented run for SDA because I'd know it could be beaten by someone just doing twice as many segments as me.

Quote:
a run with a needlessly high number of segments and low runtime punishes itself.


In terms of how people view it, true, but the trouble is that a run with fewer segments can be obsoleted by one of these insanely segmented creatures.

Quote:
This takes looking at it more than one segment at a time imo. If you segment one part in order to pull off a 20-second timesaver, with 4 more parts that were simply segmented for better optimization your timesaver is already nullified with a 4 second penalty on EACH segment.


Uh, so then you'd segment before the trick and nowhere else, allowing you to do all the tricks but not just segment needlessly to optimise running around. Which is exactly the result I'd want to see.

Quote:
I think the rules pages spell it out enough for me to understand what an accurate timer should be. ... If it stops during time where SDA would normally time, most notably dialogues and inventory screens, don't use it.


The trouble is that the rules don't say that and in practice that's not how they work. MM9 is not an exception to this, it's the norm.

Quote:
the less individual rulings for games there are, the better.


This has come up frequently in this thread. I don't really get it - why is this so? Different games have their own issues, common sense and an idea of the spirit of the site is a better guide than rigid rules that may produce perverse and ridiculous results for some games.

ZenicReverie: I would've thought that deaths vs. deathless is a sensible distinction to make for most games, though for some there isn't enough difference to justify it. I agree it should be decided on a case by case basis.

Quote:
Quote:
As for the save penalty, pretty much any game with saving anywhere has warping upon saving and loading

When it's really obvious, then it gets a penalty.


That's not really a penalty, you're just adding in the time for the missing frames. That doesn't provide a deterrant.

Quote:
This is your opinion


That is unsurprising, since I wrote it.

Quote:
All this rule does is up the time shown on the game page and in the statid and who cares about that, really.


Actually, you raise a good point here - I don't think that the segmentation penalty should be included in the time shown on statids or on the game page, that just doesn't make any sense at all. The purpose of the segmentation penalty is to provide an objective way to compare the merits of runs with different numbers of segments, but it would make more sense to calculate the times including penalties in the verification threads to decide whether a run should obsolete another, and then on the game page to write the actual time, not including penalties.

Quote:
If we'd bring even more restrictions we'd potentially prevent a new run(s) from getting on the site, also probably we'd successfully discourage some runners from contributing


Removing the segmentation penalty would stop me from contributing because what's the point in submitting a segmented run where you've not put in lots of unnecessary segments if someone can just double the segment count, copy the run and beat it by a few seconds through extra optimisation, and that will obsolete mine? When comparing runs of save-anywhere games we need to somehow capture the idea that extra pointless segments = a bad thing, and while the segmentation penalty does a pretty inadequate job of this, surely it's better than nothing? Hell, if there was no penalty I'd segment every 5 seconds just out of laziness and the desire to finish the run faster.

Edit: Also, it's not just about the possibility for obsoletion. When I'm creating a run, I want to optimise it as much as possible, and when I'm watching a run I want it to be optimised as much as possible. If there isn't some objective measure of how the 'badness' of extra segmenting weighs in against the 'goodness' of making a faster run, there's no way to decide when to segment. Having a penalty makes segmentation decisions part of the strategy of the run, and that's a good thing.
Edit history:
65: 2009-06-14 01:00:11 pm
4)
Quote from groobo:
I'll ask again: what's the goal of the site? A show of skill, or entertainment?


I've always considered this site to be primarily about skill, entertainment is something that you can have unless it restricts speed or skill. If I want to watch something that's made primarily entertainment in mind, I go to TASvideos.

Scripts are something that usually either (to my knowledge, but as far as I know one might be able to go even further with this, running whole parts of games etc.) enable you to a) perform feats with accuracy that's impossible to a human or b) make it easier to attain performance that a great player can do that the runner on the job might or might not be able to replicate or c) and easy task made automatic, allowing the player to concentrate on more difficult things at hand.

a), in my opinion, is heavily against the spirit of the site, at least how I see it. TAS' key input movies work similarly; it is a tool to break through human limitations. This site is about showing what can be done with human skill, not what could be theoretically done. Because of that my opinion is that these sort of scripts should not be allowed at all.

One might ask what counts as something beyond human limitations, of course. I think such a thing would be fairly easily characterized by things that enable the player to control extremely precise set of motions within a very tightly set time frame (and quite likely for an extended time) and/or allowing ridiculously challenging feats to be pulled off in a frequent manner. Assuming one could make scripts for console games (like a remake of some earlier game) that allows one to pull off things that shouldn't be possible to do without modifying the hardware (up+down, for example), it would also be one of the things counted as a) material.

So yes, the HL turn 180 - shoot exactly one shot - turn 180 in mere frames, frequently and always perfectly would fall into a) category in my book, especially when you consider you'd still have to bunny hop etc. during all this.

b) & c) are more a question about laziness than about doing the impossible, though b) is still on a borderline case at its "worst". However, if it enables the player to achieve better times because they free up the extra resources for the player to pull off insane maneuvers, they can be useful & inspiring. I would still recommend a separate category for such scripts.

For the record, having turbo controller runs as secondary categories is fine by me, since they'd share similar characteristics as script types b) & c).

6) Even though these sort of things do occur occasionally, I think it'd be better to keep track of both categories. To me, it's not about any% and 100% being similar, it's about having the freedom of choice in any% that one doesn't have in 100%. They just happen to have similar solutions, as far as the current knowledge goes, at least; it could change if a trick was discovered that'd alter the equation. And if you were to leave out either of the categories, which one would you ditch?

Note that when it comes to games like Mega Man 5 & 6 and Rosenkreuzstilette, where you have exactly one optional item to collect, there's very little point in tracking both as it's more about the effectiveness of two different strategies and less about collecting and backtracking.

7) I think this is fine as it is - as I've understood it, you time it with the timer as long as it's accurate and it's consistent (no dropped seconds etc.)

2) Gimme a shout once you've done with whatever changes you agree upon so I can get cracking with the Finnish translation pages.
TIOLET!
Quote:
Uh, so then you'd segment before the trick and nowhere else, allowing you to do all the tricks but not just segment needlessly to optimise running around.


There are actually examples where you need to segment just to optimize "running around". Off the top of my head I can mention map 2 in Doom 3. Even though there isn't a single hard trick in the entire map you can do it almost a minute faster using 5 segments instead of 1, and a minute is not exactly a small improvement. I wouldn't want half of that to go to waste due to some weird penalty.
Quote from Kibbo:
Quote:
Uh, so then you'd segment before the trick and nowhere else, allowing you to do all the tricks but not just segment needlessly to optimise running around.


There are actually examples where you need to segment just to optimize "running around". Off the top of my head I can mention map 2 in Doom 3. Even though there isn't a single hard trick in the entire map you can do it almost a minute faster using 5 segments instead of 1, and a minute is not exactly a small improvement. I wouldn't want half of that to go to waste due to some weird penalty.


Well, you'd still have saved 40 seconds of penalised time. The question with the segmentation penalty is how much time has to be saved for adding an extra segment in to be a good thing? Or, if the answer to that question is 'it depends', how do we determine whether or not extra segmentation is a good thing or not?