Version differences in general and how we should treat them is an interesting topic.
In some cases, it's obvious that two versions are completely different games and should be published separately. A good example of this are some remakes like SM64DS. Then we have updated ports that are mostly interesting on their own based on their additional content (I'll go with another Mario example, the "You vs. Boo" TAS of SMB Deluxe).
It gets more difficult when we get to things like the countless ports of Doom or 1.1 releases. SMB Deluxe did eventually get an any% publication because it was deemed different enough. But it's harder to find examples of separate publications of closer variations of the same game (SMB Deluxe is still on a different platform after all!).
I tried to think of some examples that would fall into such a category, versions of the same game that are very close (not a full remake or significantly different platform), but have some notable differences. One game that comes to my mind is Tenchu: Stealth Assassins with its separate publications of JP and U. But those two TASes are both in the moons tier. What would have happened if one of the TASes wouldn't have been deemed entertaining enough for moons, would it have been rejected?
Let me bring up a game that hasn't been TASed yet because of technical limitations:
Sly Cooper and the Thievius Raccoonus was released for PS2 and later got an HD port for PS3 as part of the Sly Trilogy. The PS2 version has all the glitches and is very entertaining to watch as a speedrun, especially since many of the non-gameplay parts can be skipped in some way. The PS3 port removes the biggest and most spectacular skips and also forces you to watch all those cutscenes that you could skip in the original. From a technical perspective, the run is significantly different between the two releases: The original is fast paced and can skip 90% of two of the games chapters, while the port has to play every single level and is mostly reduced to use skips within the regular gameplay. Would a TAS of the PS3 version get rejected for having less entertainment value than the original, despite being a very different game? Or would it still get a vault publication because it's different enough?
If there is an example of two versions of the same game being released in moons and vault, let me know!
Current project: Gex 3 any%
Paused: Gex 64 any%
There are no N64 emulators. Just SM64 emulators with hacky support for all the other games.
I'd personally prioritize:
... Version preferred by majority of audience ...
The following suggested criteria I would not consider at all not even one iota:
... Version popularity/sales ...
What do you see as the difference between the most preferred version and the most popular version?
Joined: 3/9/2004
Posts: 4588
Location: In his lab studying psychology to find new ways to torture TASers and forumers
What our audience prefers for TASs is not necessarily related to what is the most popular version.
Warning: Opinions expressed by Nach or others in this post do not necessarily reflect the views, opinions, or position of Nach himself on the matter(s) being discussed therein.
Beyond that we have different versions of games where input will never sync between the two, not even in limited (but complete) segments. For those, I don't see any parity between the games anymore, despite a similar look and feel, and it being common knowledge that the two are supposedly the same.
That's a very interesting possible approach at the question of "should these versions of the game be considered the same game?" If the exact same optimal TAS input works for both (or at least in the actual playable parts of both), they would be considered the same game, but if an optimal TAS input for one would never sync with the other (even with possible delays added or removed from the input), then they could be technically considered different games.
(That still wouldn't mean that both should be automatically accepted. There still ought to be some notable differences between the two that justifies publishing both. But the minimum requirement of "it's a different game" would at least have been fulfilled by this.)
By the way, a couple of questions I thought of, for your consideration:
1) If the site were redesigned to be more like speedrun.com, as I suggested earlier, with each game having its own page, with all the TASes of said game being neatly and clearly listed or tabulated there, would you agree that this would perhaps allow loosening the principle of keeping the number of branches/categories for a single game minimal (since perceived clutter wouldn't be a problem anymore)?
2) If you answered the previous in the positive, do you think it would be a silly idea to start the loosening of that principle now, even prior to such a potential site redesign? Should we just wait for the redesign to happen before we start accepting more categories for games?
I agree that a submission should not focus only on speed if it is to be considered for higher tiers, but that's not the point I'm trying to make. My point is that PAL runs shouldn't be under any higher pressure to deliver that NTSC runs, independent of the tier. Just because the new sumbission happens to be on the PAL version, it shouldn't be held to higher standards or have something expectional in it, a simple improvement (considering all aspects) is good enough. Again, all the usual caveats apply, they should be good versions and provide similar experiences.
Using the case-by-case based approach doesn't add any extra pressure. If the audience and the judges find the PAL version worth a replacement, it is justified. Replacing regardless of people's opinions just because something is faster is questionable exactly because of that: Why are we even doing it that way if it's not what people prefer?MrWint wrote:
Maybe think of it the other way. If there is an already published PAL version, and someone tried to obsolete it with an NTSC version (and the NTSC version happened to be released first), would you expect the same level of strict superiority and strictly better media, gameplay, difficulty, challenge, glitches, routes, etc?
Whenever the question is "Should we switch to another region version instead of having both?", the answer should be "Yes, if the version to switch to is in some aspects better". Otherwise we won't be improving our overall content.MrWint wrote:
If you wouldn't expect that, it points to the bias for NTSC and original versions that I'm arguing should be dropped. If you would expect it, then you're creating an environment where whatever version got published first has a large advantage, where it's easy to publish improvements on that version but not any different one, regardless of any other factors. This seems arbitrary to me.
You're asking me a Yes or No question and telling me that both answers are flawed? I'd say the question is flawed :P
andypanther wrote:
Let me bring up a game that hasn't been TASed yet because of technical limitations:
Sly Cooper and the Thievius Raccoonus was released for PS2 and later got an HD port for PS3 as part of the Sly Trilogy. The PS2 version has all the glitches and is very entertaining to watch as a speedrun, especially since many of the non-gameplay parts can be skipped in some way. The PS3 port removes the biggest and most spectacular skips and also forces you to watch all those cutscenes that you could skip in the original. From a technical perspective, the run is significantly different between the two releases: The original is fast paced and can skip 90% of two of the games chapters, while the port has to play every single level and is mostly reduced to use skips within the regular gameplay. Would a TAS of the PS3 version get rejected for having less entertainment value than the original, despite being a very different game? Or would it still get a vault publication because it's different enough?
We don't obsolete games between different consoles unless:
- the consoles are very close to each other (like GG and SMS), the games are close to each other, and the audience agrees with obsoletion
- game versions are intentionally made (almost) identical, like it uses to happen with modern game versions for consoles of the same generation
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.
Joined: 3/9/2004
Posts: 4588
Location: In his lab studying psychology to find new ways to torture TASers and forumers
Warp wrote:
If the exact same optimal TAS input works for both (or at least in the actual playable parts of both), they would be considered the same game, but if an optimal TAS input for one would never sync with the other (even with possible delays added or removed from the input), then they could be technically considered different games.
We do have the possible interesting case where best the TAS for game revision B syncs on game revision A, but the best for game revision A does not sync on game revision B because some bug was fixed.
Warp wrote:
1) If the site were redesigned to be more like speedrun.com, as I suggested earlier, with each game having its own page, with all the TASes of said game being neatly and clearly listed or tabulated there, would you agree that this would perhaps allow loosening the principle of keeping the number of branches/categories for a single game minimal (since perceived clutter wouldn't be a problem anymore)?
I have no familiarity with this other site.
Even if TASs are organized by game, if you see 30 TASs on a page for a single game, and what differentiates them isn't clear could be difficult for viewers.
feos wrote:
We don't obsolete games between different consoles unless:
- the consoles are very close to each other (like GG and SMS), the games are close to each other, and the audience agrees with obsoletion
- game versions are intentionally made (almost) identical, like it uses to happen with modern game versions for consoles of the same generation
We've had a few exceptions to this where a poor version was obsoleted by a better one (as was originally done for your Battletoads and Double Dragon). We've also used the technique to get rid of really old TASs that in retrospect we don't think should have been published in the first place.
Warning: Opinions expressed by Nach or others in this post do not necessarily reflect the views, opinions, or position of Nach himself on the matter(s) being discussed therein.
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.
I don't think it matters how large their differences technically are. It may be a 1.0 vs. 1.1 version of the otherwise same title, or it may be a port for a completely different system altogether, with a different underlying engine and everything. The important point is that they have the same goals, use similar means to achieve them, and provide a similar viewing experience
I think how large the differences technically are is the primary crucial point, and is in fact a foundation of TASing philosophy. Please see my above response to Warp and link.
I need to clarify here, I agree that the technical details and differences are very important when making a TAS (I hadn't created the SMB disassembly if I thought the techincal differences are not important). What I was talking about is the way I perceive TASVideos chooses categories and which runs potentially share a slot and compete with each other for publication. If you have a straight SNES remake of a NES game, then iiuc they usually compete for a slot, at most one is published to avoid redundancy. The PAL SMB run was rejected not because it was objectively bad, but because it competed with the existing NTSC publication and lost.
Nach wrote:
Two runs for a single game or two similar games each with their own run? I assume you mean the latter, as the former makes little sense in the context.
This gets into a number of issues regarding quality of the game, TAS itself, popularity, audience response. I think most of the time we should strive for both. If they're too similar and absolutely cannot do both, I'd personally prioritize:
Highest quality game version (most potential for TASability and good audience feedback)
Version preferred by majority of audience
Audiovisual quality of the version
I define TASability as including ability to complete the games in multiple ways, many tradeoffs which require study and planning, much depth that has a near infinite supply of finding new things, ways to perform entertaining action often.
The following suggested criteria I would not consider at all not even one iota:
Fastest time
Version popularity/sales
Which version was published first on TASVideos
Which version was released first to market
I wouldn't consider any of them because fastest time for some run in some branch, most sales, first to publish, and first to market have 0 affect on the game's quality or its overall TAS potential.
Ok, I need to apologize for conflating two concepts into one, comparing two versions of a game and comparing two runs (submissions) of a game (possibly using different versions). Also, I mixed version popularity and preferred by majority of audience, I agree that your nomenclature makes more sense and it reflects what I actually meant when I said version popularity (and adding "sales" as an afterthought didn't make it any better).
However, I want to challenge your statement that fastest time is not a factor at all. When considering the Pokémon Gen1 SRAM movie, which used Red to obsolete Yellow, I'd argue that time was a major factor. The run features no major new tricks, and is admittedly a downgrade visually from GBC to GB, so you could argue it used an inferior version of the game. Of course this run is almost exclusively based on technical details and the entertainment value is limited anyway. The audience response was pretty positive, but that is only because it was faster.
I can see the argument that the fastest time is only a means that as no impact by itself and the audience response is what actually matters (like game sales don't mean anything by themselves but usually drive audience preferences which do matter). On the other hand you promote time as a factor heavily on TASVideos, when I visit the home page of TASVideos, the information I'm presented with for all the listed runs is the name of the game, the category, and the time, so in a way you are promoting time as a driver for audience response.
How time is factored into the judgement is actually besides the point I'm trying to make, so I'll stop rambling about it. I'm interested in how you reconcile your statement that which version was released first to market has no impact with your statements on the originality of NTSC SMB as a reason to prefer it over the PAL version (e.g. here), as well as the Movie Rules which state that PAL ROMs are generally not allowed unless there are significant merits.
Reading the rules in context of our discussions so far, I actually don't think it's unreasonable to state a general preference for NTSC ROMs, but this is only on the basis that they are likely to have a better audience response (because of their popularity), and that PAL ROMs tend to be broken because of the compensation for different clock speeds. Disallowing them in general unless special conditions are met is a step too far in my opinion. I think this could be fixed just by rephrasing the section, explaining that PAL ROMs are in fact allowed, but all the caveats apply so the NTSC version is the better choice in most cases. The current wording sends the wrong signals, and it causes rulings based on rules that should be guidelines.
feos wrote:
MrWint wrote:
I agree that a submission should not focus only on speed if it is to be considered for higher tiers, but that's not the point I'm trying to make. My point is that PAL runs shouldn't be under any higher pressure to deliver that NTSC runs, independent of the tier. Just because the new sumbission happens to be on the PAL version, it shouldn't be held to higher standards or have something expectional in it, a simple improvement (considering all aspects) is good enough. Again, all the usual caveats apply, they should be good versions and provide similar experiences.
Using the case-by-case based approach doesn't add any extra pressure. If the audience and the judges find the PAL version worth a replacement, it is justified. Replacing regardless of people's opinions just because something is faster is questionable exactly because of that: Why are we even doing it that way if it's not what people prefer?
That's not at all what I meant. When I said "improvement", it may not only mean a faster time, but can also mean better audience response, technical excellence, or other things. Just a better run in whichever ways is appropriate for the category and tier, just like an improvement using the same game version would.
feos wrote:
MrWint wrote:
If you wouldn't expect that, it points to the bias for NTSC and original versions that I'm arguing should be dropped. If you would expect it, then you're creating an environment where whatever version got published first has a large advantage, where it's easy to publish improvements on that version but not any different one, regardless of any other factors. This seems arbitrary to me.
You're asking me a Yes or No question and telling me that both answers are flawed? I'd say the question is flawed :P
Actually no, logic dictates that when you are left with only flawed answers, it's not the question that is flawed but one of the assumptions you made along the way. You can see the question as rhetorical, it was a tool to make my point that having a double standard for NTSC and PAL creates these impossible situations, thus showing the premise itself must be flawed.
I'm still somewhat interested in an answer to the question actually, in order to get to the bottom of our disagreement, because either you actually disagree with the premise yourself and this is just a misunderstanding, or you are disagreeing with one of my interpretations of the results I described as flawed, and see this as a preferable outcome instead, which hopefully leads to more insights into the nature of our disagreement. I could have asked the question without giving my interpretation of the answers straight away, but I figured this is the more efficient way of communication given the roundtrip time. If this appeared mocking or demeaning to you, this was not at all my intention.
Joined: 3/9/2004
Posts: 4588
Location: In his lab studying psychology to find new ways to torture TASers and forumers
MrWint wrote:
Nach wrote:
I think most of the time we should strive for both. If they're too similar and absolutely cannot do both, I'd personally prioritize:
snip
I wouldn't consider any of them because fastest time for some run in some branch, most sales, first to publish, and first to market have 0 affect on the game's quality or its overall TAS potential.
However, I want to challenge your statement that fastest time is not a factor at all. When considering the Pokémon Gen1 SRAM movie, which used Red to obsolete Yellow, I'd argue that time was a major factor. The run features no major new tricks, and is admittedly a downgrade visually from GBC to GB, so you could argue it used an inferior version of the game. Of course this run is almost exclusively based on technical details and the entertainment value is limited anyway. The audience response was pretty positive, but that is only because it was faster.
I can see the argument that the fastest time is only a means that as no impact by itself and the audience response is what actually matters (like game sales don't mean anything by themselves but usually drive audience preferences which do matter). On the other hand you promote time as a factor heavily on TASVideos, when I visit the home page of TASVideos, the information I'm presented with for all the listed runs is the name of the game, the category, and the time, so in a way you are promoting time as a driver for audience response.
As you pointed out, time can be a factor in how an audience feels, but it is not the main concern. Regarding how TASVideos presents data there, it's not my personal preference.
MrWint wrote:
How time is factored into the judgement is actually besides the point I'm trying to make, so I'll stop rambling about it. I'm interested in how you reconcile your statement that which version was released first to market has no impact with your statements on the originality of NTSC SMB as a reason to prefer it over the PAL version (e.g. here), as well as the Movie Rules which state that PAL ROMs are generally not allowed unless there are significant merits.
My personal preferences are not in alignment with the movie rules.
I'd also point out towards your earlier point, that if speed of games are being considered for audience preference without getting into the particulars of any given run, you need to factor in all branches (such as stuff like walkathon and warpless in your example case). Which increases the amount of technically different factors to consider in determining which is faster, which can make different branches have different outcomes, sometimes non-intuitively.
MrWint wrote:
Reading the rules in context of our discussions so far, I actually don't think it's unreasonable to state a general preference for NTSC ROMs, but this is only on the basis that they are likely to have a better audience response (because of their popularity), and that PAL ROMs tend to be broken because of the compensation for different clock speeds. Disallowing them in general unless special conditions are met is a step too far in my opinion. I think this could be fixed just by rephrasing the section, explaining that PAL ROMs are in fact allowed, but all the caveats apply so the NTSC version is the better choice in most cases. The current wording sends the wrong signals, and it causes rulings based on rules that should be guidelines.
I think a key point people are missing is that with the way the rules are structured currently, games which have technical differences but have a very similar look and feel will basically always prefer NTSC unless there is something immensely significant altering things in favor of PAL.
However, if our sole aim is looking to reword things for greater clarity of what the existing rules are without actually changing them, I can take care of that.
Warning: Opinions expressed by Nach or others in this post do not necessarily reflect the views, opinions, or position of Nach himself on the matter(s) being discussed therein.
Reading the rules in context of our discussions so far, I actually don't think it's unreasonable to state a general preference for NTSC ROMs, but this is only on the basis that they are likely to have a better audience response (because of their popularity), and that PAL ROMs tend to be broken because of the compensation for different clock speeds. Disallowing them in general unless special conditions are met is a step too far in my opinion. I think this could be fixed just by rephrasing the section, explaining that PAL ROMs are in fact allowed, but all the caveats apply so the NTSC version is the better choice in most cases. The current wording sends the wrong signals, and it causes rulings based on rules that should be guidelines.
I fully agree with this. At first there was no ruling, some PAL runs just existed. Then it became "generally not preferred". And then "generally not allowed" for whatever reason.
Post #457965MrWint wrote:
logic dictates that when you are left with only flawed answers, it's not the question that is flawed but one of the assumptions you made along the way
it was a tool to make my point that having a double standard for NTSC and PAL creates these impossible situations, thus showing the premise itself must be flawed
I dislike the current ruling and suggest my own, by which NTSC and PAL are conceptually equal and have equal chances. Still, one happens to be the default version, and when you want to switch from it, occasionally or broadly, you have to have solid reasons. I seem to agree with Nach's reasons. The switch should just be legalized.
I'm still somewhat interested in an answer to the question actually, in order to get to the bottom of our disagreement
The value NTSC has is the same as left+right trick has. It's traditional. Left+right is impossible on standard controllers for some consoles or arcade machines. And some people would disagree with allowing that trick. But it's traditionally there, so if you want to change that tradition, that's where you need all your arguments, not in an imaginary case where the shift has already happened and we're just observing the outcome. And personally I don't care about what version was released first, neither do judge guidelines.
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.
I have no familiarity with this other site.
Even if TASs are organized by game, if you see 30 TASs on a page for a single game, and what differentiates them isn't clear could be difficult for viewers.
Take, for instance, the speedrun.com page for Ocarina of Time. It has a whopping 13 categories. I don't think the page is confusing at all. The different categories are neatly in labeled tabs, and the description of each category can be found in each tab (by clicking the "view rules" button).
Case by case basis seems fine to me. Maybe just change this:
Due to this, PAL versions of ROMs are generally not allowed, unless there are significant technical and/or entertainment merits to using this version.
to this:
Due to this, PAL versions of ROMs are generally not allowed, unless there are significant technical and/or entertainment merits to using this version, or there is strong audience support.
To keep things simple and consistent, just publish PAL seperately.
I'd don't think any rule should be tied to the tier system.
Due to this, PAL versions of ROMs are generally not allowed, unless there are significant technical and/or entertainment merits to using this version, or there is strong audience support.
Audience support alone doesn't prove anything, if it can't be formalized in some statement, that is then put as a reason to accept a PAL submission. And to come up with such statement, we impose factors we want to get evaluated.
For vaultable runs (any% and 100%) I impose superiority in order to obsolete such run. For other branches I impose diversity as the main requirement. If these are met, PAL submissions should get published without any pressure.
To me the new question is... How many people think we should not change the movie rules, nor clarify them?
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.
Audience support alone doesn't prove anything, if it can't be formalized in some statement, that is then put as a reason to accept a PAL submission. And to come up with such statement, we impose factors we want to get evaluated.
To me the new question is... How many people think we should not change the movie rules, nor clarify them?
Well yeah honestly that's kind of what I was going for. It doesn't really need to prove anything except that people like it. And if we just want to be able to publish the couple of good PAL runs that come along every once in a while, just assert that we can do so.
It's kind of like Mothrayas' "special one time exception." If it's good enough to have, just say it's ok. (And I mean that sincerely, not sarcastically, exceptions are needed every now and then.) In this case it's just slightly more formalized since many PAL ports really aren't worth having. But, some are (like PAL SMB in my opinion), so let's leave an opening for them.
As for your bolded question, I really do think something needs to change here. We shouldn't be turning away good runs like SMB PAL was when the site is littered with horribly boring content that's accepted just because someone was able to get a game on a cart and sell it.
It's kind of like Mothrayas' "special one time exception." If it's good enough to have, just say it's ok. (And I mean that sincerely, not sarcastically, exceptions are needed every now and then.) In this case it's just slightly more formalized since many PAL ports really aren't worth having. But, some are (like PAL SMB in my opinion), so let's leave an opening for them.
The problem with SMB PAL is that neither superiority nor diversity can be formalized for it. "One time exceptions", as moozooh pointed out, should either be legalized or not happen at all. My goal is to make the PAL ruling fully transparent, with no surprises. If we have to make an exception every now and then, the ruling is sloppy.
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.
Which is why I re-propose using the fastest version regardless of what version it is or what region it is (once you take into account that which TASVideos.org already excludes - title screen, language change, and cutscenes). If the fastest version is not acceptable for some reason, it's on the site to show why that version is not good enough, not on the submitters to show why their version is good enough.
Enough with region-specific rules. Pick whatever version you want, regardless of region (preference on North America in cases where fastest is identical across multiple regions), as long as that version is an acceptable version (i.e. it's not a version that current rules for TASVideos.org would reject outright regardless of region).
"Use the fastest version available as an official release. In the event that multiple versions are identical, after discounting changes due to region (title screen length, language change), preference is for North American versions, but this is not required."
I understand preference for "original" and stuff, but in reality, we're getting more and more mixed with actual speedruns, and yet our hard stance on versions has been one of the main detractors. We had a recent rejection of a TAS which is faster in every way because it is done on an "inferior version" than the current TAS. That is ludicrous, and needs to change.
Which is why I re-propose using the fastest version regardless of what version it is or what region it is (once you take into account that which TASVideos.org already excludes - title screen, language change, and cutscenes). If the fastest version is not acceptable for some reason, it's on the site to show why that version is not good enough, not on the submitters to show why their version is good enough.
And what is the very reason to switch to that approach? Because Habreno said us to? Here are my reasons not to blindly grab the fastest version:
feos wrote:
If the audience and the judges find the PAL version worth a replacement, it is justified. Replacing regardless of people's opinions just because something is faster is questionable exactly because of that: Why are we even doing it that way if it's not what people prefer?
feos wrote:
I dislike the current ruling and suggest my own, by which NTSC and PAL are conceptually equal and have equal chances. Still, one happens to be the default version, and when you want to switch from it, occasionally or broadly, you have to have solid reasons. I seem to agree with Nach's reasons. The switch should just be legalized.
The value NTSC has is the same as left+right trick has. It's traditional. Left+right is impossible on standard controllers for some consoles or arcade machines. And some people would disagree with allowing that trick. But it's traditionally there, so if you want to change that tradition, that's where you need all your arguments, not in an imaginary case where the shift has already happened and we're just observing the outcome.
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.
I'd also point out towards your earlier point, that if speed of games are being considered for audience preference without getting into the particulars of any given run, you need to factor in all branches (such as stuff like walkathon and warpless in your example case). Which increases the amount of technically different factors to consider in determining which is faster, which can make different branches have different outcomes, sometimes non-intuitively.
It sounds like you're implying that the preferred version of a game should be determined independently of the branches. I disagree, I think it's entirely reasonable for different branches to have different preferred versions, so there's no need to factor in all branches when making these decisions. E.g., the SMB warped category I think PAL is a reasonable alternative to NTSC, whereas for warpless you'd lose a lot of entertainment value by cutting out the flag pole glitch setups.
Nach wrote:
MrWint wrote:
Reading the rules in context of our discussions so far, I actually don't think it's unreasonable to state a general preference for NTSC ROMs, but this is only on the basis that they are likely to have a better audience response (because of their popularity), and that PAL ROMs tend to be broken because of the compensation for different clock speeds. Disallowing them in general unless special conditions are met is a step too far in my opinion. I think this could be fixed just by rephrasing the section, explaining that PAL ROMs are in fact allowed, but all the caveats apply so the NTSC version is the better choice in most cases. The current wording sends the wrong signals, and it causes rulings based on rules that should be guidelines.
I think a key point people are missing is that with the way the rules are structured currently, games which have technical differences but have a very similar look and feel will basically always prefer NTSC unless there is something immensely significant altering things in favor of PAL.
I think you hit on the main issue I see with the current rules here. For games which have technical differences but have a very similar look and feel, PAL doesn't need any "immensely significant altering things" in order to be favored, it just needs to be preferred considering all the aspects we discussed already. I don't think that will change the outcomes of most decisions, the NTSC version will be the preferred version for almost all games, but that is not because it is the "default" version, but becasue it wins out against the other versions when comparing the relevant criteria.
In the judge's notes of the SMB PAL TAS, I expected to find such a comparison of the games for this game and this branch specifically, but they completely skipped that and only discussed a separate category as an option, giving the impression (reinforced by the infamous "lunacy" quote) that there was no debate and NTSC is just the default version. I didn't expect the PAL version to win, but I expected it to be considered.
My proposal to remedy this is to avoid statements like "PAL versions of ROMs are generally not allowed" in the Movie Rules, and rather state the aspects that need to be considered, and possible caveats of using PAL versions (most of these things are mentioned already). Since NTSC will be the winner in that comparison for most games and branches, stating a general preference for NTSC is useful to guide people to using the right version without forcing them to do in-depth comparisons, but only as a guideline, not as a rule.
feos wrote:
MrWint wrote:
logic dictates that when you are left with only flawed answers, it's not the question that is flawed but one of the assumptions you made along the way
You can't claim my statement is sophistry without any arguments why is it flawed or why the question doesn't make sense in that context. Questions with only impossible answers are not automatically a sophism. I maintain that this is a perfectly valid proof by contradiction. Our argument itself has been resolved in my eyes though.
Joined: 3/9/2004
Posts: 4588
Location: In his lab studying psychology to find new ways to torture TASers and forumers
MrWint wrote:
Nach wrote:
I'd also point out towards your earlier point, that if speed of games are being considered for audience preference without getting into the particulars of any given run, you need to factor in all branches (such as stuff like walkathon and warpless in your example case). Which increases the amount of technically different factors to consider in determining which is faster, which can make different branches have different outcomes, sometimes non-intuitively.
It sounds like you're implying that the preferred version of a game should be determined independently of the branches.
I'm implying that's an option based on the discussion so far. In discussions about rule changes, unless you consider exactly what it is that you're trying to define, you end up with these questions.
MrWint wrote:
I think you hit on the main issue I see with the current rules here. For games which have technical differences but have a very similar look and feel, PAL doesn't need any "immensely significant altering things" in order to be favored, it just needs to be preferred considering all the aspects we discussed already.
Preferred in what context? You must be clear about this.
For acceptance? For obsoletion?
For the game as a whole? For individual branches?
MrWint wrote:
In the judge's notes of the SMB PAL TAS, I expected to find such a comparison of the games for this game and this branch specifically, but they completely skipped that and only discussed a separate category as an option, giving the impression (reinforced by the infamous "lunacy" quote) that there was no debate and NTSC is just the default version. I didn't expect the PAL version to win, but I expected it to be considered.
You're confusing two things here. PAL was considered for acceptance. It was not seriously considered for obsoletion because the majority of judges involved didn't see any sane way the two are comparable, hence, it's pure lunacy.
Warning: Opinions expressed by Nach or others in this post do not necessarily reflect the views, opinions, or position of Nach himself on the matter(s) being discussed therein.
In the scenario we're discussing (games which have technical differences but have a very similar look and feel) I think acceptance and obsoletion end up being the same. Since introducing another branch would create unwanted redundancies, the only way it gets accepted is by obsoleting the existing movie (if any). And as mentioned before, I only consider that specific branch, not the game in general (other branches may have different outcomes).
You're not really doing any better in terms of clarity though, regarding comparability, what does it mean? Comparing time? Audience response? Version quality?
Having versions that are not comparable (assuming it means that there's no way to choose a better or worse version out of the two) in the same publication slot is not very satisfying.
In this case, it seems you decided to stick with the currently published version. But does the version that is published first always win? If the currently published version were PAL and the new submission NTSC, would you stick to the published PAL version then?
I think there should be a preferred version in all cases where two runs end up in the same publication slot, and I'm trying to formalize what this decision would be based on. If we absolutely can't find a way, it's a sign they should be separate categories then, to avoid that awkward middle ground.
The problem with SMB PAL is that neither superiority nor diversity can be formalized for it. "One time exceptions", as moozooh pointed out, should either be legalized or not happen at all. My goal is to make the PAL ruling fully transparent, with no surprises. If we have to make an exception every now and then, the ruling is sloppy.
I think it's ok if you can't formalize those two things. It can be published alongside SMB NTSC just fine on it's own merits (people just saying they liked it.)
I also think exceptions are fine. After all, if judging were nothing more then linearly applying the rules, you wouldn't need to call yourselves judges anymore, I don't think that's an atainable goal at any rate.
But it seems we have different goals here in how to sort this out.
MrWint wrote:
In the scenario we're discussing (games which have technical differences but have a very similar look and feel) I think acceptance and obsoletion end up being the same. Since introducing another branch would create unwanted redundancies, the only way it gets accepted is by obsoleting the existing movie (if any). And as mentioned before, I only consider that specific branch, not the game in general (other branches may have different outcomes).
I personally don't see the clutter argument as a real concern. People can find both NTSC and PAL entertaining, and nothing bad will happen by having them published side by side (again, this assumes that the PAL port in quesiton is sufficiently well done like SMB.)
Well, I don't think I can add much more to this though. I don't see the need, or the practically, of having a strict formalism here if the goal is to allow some PAL runs while still preferring NTSC. Saying it's ok and being done with it seems good enough to me. If others like feos are trying to really formalize things here then I wish them luck, as the whole situation is quite messy (as the real world usually is), and I think the end result would end up very little different then just using an exception, probably with just many more words.
The problem with SMB PAL is that neither superiority nor diversity can be formalized for it. "One time exceptions", as moozooh pointed out, should either be legalized or not happen at all. My goal is to make the PAL ruling fully transparent, with no surprises. If we have to make an exception every now and then, the ruling is sloppy.
I think it's ok if you can't formalize those two things. It can be published alongside SMB NTSC just fine on it's own merits (people just saying they liked it.)
I also think exceptions are fine. After all, if judging were nothing more then linearly applying the rules, you wouldn't need to call yourselves judges anymore, I don't think that's an atainable goal at any rate.
And vice versa. If we just go by the audience no matter how silly their opinion is, how much it contradicts with all the judges' experience, we aren't judges either. I used to absolutize audience once, I know what I'm talking about.
If you can't formalize it, you can't build a consensus around it. Also it's quite common that the audience is misleading: they can give dozens Yes votes in the submission thread and ecstatically express how much they loved the movie, and when we publish it in Moons, it gets 3/10 by the vast majority. Happened to me as well.
If we even require that branches need to be different from each other, we would end up with this:
- Was it different enough for a new branch?
- Yes!!!
- What exactly was different?
- ...............
- But you're certain it was different?
- Yes!!!
That basically happened when Nach was interviewing the judges. He didn't ask the audience to exactly formalize the difference, but I bet everyone would fail too.
And if you don't require difference and just publish what they like, their taste will quickly degrade, along with our overall content quality. There will be this mad child that gets hysterical if you don't do what it wants, even if it can't tell you what's so good about it. Or worse, several mad children each of them wanting the opposite. We don't want that.
Alyosha wrote:
I don't see the need, or the practically, of having a strict formalism here if the goal is to allow some PAL runs while still preferring NTSC.
This is what we need to retire. We should not prefer NTSC, we should just end up with it being the primary version even after giving PAL a chance, like MrWint says. What we need to prefer is superiority.
Alyosha wrote:
I personally don't see the clutter argument as a real concern. People can find both NTSC and PAL entertaining, and nothing bad will happen by having them published side by side (again, this assumes that the PAL port in quesiton is sufficiently well done like SMB.)
One important part of the clutter argument is that it increases the work everyone should do just because people happen to like something. Look at the current queue. Submissions arrive at the same rate as we judge and publish them. If we have one less publisher or one less judge, they will start jamming. We don't want that. So we only want to publish things that are really worth publishing.
And to know what is worth publishing, we have those rules and guidelines. They also help us maintain the quality level of our publications. And diversity is an important part of that, because then the viewer can get an idea of what to expect. And finally, elaborately filtering stuff we publish has been one of the basic traditions here. Having high entry barrier helps to have better content than otherwise.
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.
Joined: 3/9/2004
Posts: 4588
Location: In his lab studying psychology to find new ways to torture TASers and forumers
MrWint wrote:
You're not really doing any better in terms of clarity though, regarding comparability, what does it mean?
See what I wrote to Warp earlier.
MrWint wrote:
Comparing time? Audience response? Version quality?
Comparing input. Without a large degree of input compatibility, it's a different game.
Warning: Opinions expressed by Nach or others in this post do not necessarily reflect the views, opinions, or position of Nach himself on the matter(s) being discussed therein.
I don't think the question is whether it is a different game or not. Suppose it is a different game by that definition, like in the SMB case. How do you decide which gets published now? You decided against publishing both, so you need to decide for one or the other. That decision is what I'm interested in, and the criteria used to decide it.
