The True Max & Max Exceptions Debate

4shockblast said:

With respect to the deaf imp discussion for map 10 a few pages ago, it is possible to kill all of them with no luck involved. One is alertable by splash and one by chaingun fire as shown in Bloodite's max. The one he missed can be alerted by running off a nearby platform at the right spot with enough speed and he will see the player. As far as I can tell, there is no excuse not to go for a valid max here. :V

^ Yet 100% is still not required for pl27, lv27 or ev15 when all of these can be done with simple setups.

So, will tysons of those maps now become invalid and undoable? Or are we going to cherrypick and keep the old rules for special cases?

It's debatable, but I don't think it will change in CN rules anyways. Globally speaking, I'd say having to compensate for these mapping errors should be decided on a case by case basis. Most importantly, if the solution radically increases difficulty, luck factor, ridiculous detouring or it just straight up requires an ugly demo, then I'd say no, it's just not worth it and an exception should be kept. For example, I believe it was memfis who once found a map that had an extra pointless secret sector that'd require a free glide for triggering it, screw that sort of hassle. Then again if there's indeed a simple setup that reliably allows you to kill more monsters than "required", then it should absolutely obsolete the old demos. Afaik there even was a "precedent" a few years ago when tatsurd found a way to kill some "unkillable" mancos in a mm map (but I'm winging it because I gotta run soon and don't have the time to fact-check, heh).

I agree with dew. Need to discuss every case separately. If there doesn't seem to be a reliable solution, or if it requres waiting like minutes for a monster to show up, then in most cases you don't need 100% kills I think.

dew said:

For example, I believe it was memfis who once found a map that had an extra pointless secret sector that'd require a free glide for triggering it, screw that sort of hassle.

Yeah, here it is if someone wants to see it: http://doomedsda.us/wad1624.html Ride the lowering platform at the start, open the door and press on the third lamp to the left.

If you don't want to kill an enemy or get a secret because it's dumb, that's fine. But don't pretend that it's still max. Just accept the fact that you don't want to max the map, and you are running 99%, and if people want to compete on 99%k then that's fine, let them compete, but don't just put on a blindfold and say "this is max" when it isn't.

Some runs have stupid tricks, and it's more fun to run/compete on a trickless route, which is fine, but I'm not going to tell someone not to run with the trick just because I want my run to be at the top of the list.

Whether this is treated case-by-case or not, I would argue that killing all the imps on su10 is not too difficult or dumb to do. It's a relatively small part of the map, requires little to no luck, and is not difficult to perform.

I do believe it needs to be case-by-case at least to handle stuff like Plutonia map 11 exit archviles and Scythe map 28, but the question of what monsters are too ridiculous to kill or secrets too ridiculous to obtain for max is debatable and obviously changes over time, so it's hard to create a concrete set of rules.

su10 is a perfect example of a simple fix that should be implemented to remove those imps. there are only 4 max demos on the dsda - and then we wont have to have this stupid argument !

kraflab said:

If you don't want to kill an enemy or get a secret because it's dumb, that's fine. But don't pretend that it's still max. Just accept the fact that you don't want to max the map, and you are running 99%, and if people want to compete on 99%k then that's fine, let them compete, but don't just put on a blindfold and say "this is max" when it isn't.

That's your definition of a max. You can't really just pry "reasonability" out of the equation and leave it at that. E1M8, PL11, etc.

rehelekretep said:

su10 is a perfect example of a simple fix that should be implemented to remove those imps. there are only 4 max demos on the dsda - and then we wont have to have this stupid argument !

No, fuck that. An unofficial "simple fix" is eternally more hassle than just agreeing upon something, because you're indirectly creating more mess with all the desyncing and incomparabilities to previous times. I can swallow a fix when a map lacks exit trigger, but every other case should just get bent.

ʢꗞ ͟ʖꗞʡ

No matter how "simple" the fix is you can't neglect the fact that it takes 40 mins to max this map. By no means is this a simple max either...

Beyond just being a hassle and leading to desyncs, map fixes aren't always possible to account for such cases because it can be hard to determine intent. Clearly, the monsters at the end of e1m8 or Plutonia map11 or pl2 map 11 or the zombiemen on sc28 were not intended to be killed, so you can't just fix those maps to make them maxable. Then, there's the question of map authors not always allowing edits to the maps at all (not sure if that should be considered relevant for speedrun-related fixes). Even just adding an exit to a map is questionable in my opinion. I would say that should only be done if it is clear where an exit should be (for instance, if the map has an exit switch that just doesn't work like qudoom.wad) or if a map can be treated as a boss encounter like wow.wad or kspider.wad and moved to a boss map slot. I think fixes probably should be limited to making maps completable or playable if there is an obvious intended exit that is for whatever reason unusable and preferably done if the map authors allow map edits or better without any map edits as in the case of h_phobia map 33 whenever this is possible.

Regarding the general discussion in this thread - after some discussion in the Discord channel, we came to the conclusion that it would be best to keep two separate categories. First, there could be a "true max" category that may not always be possible but that would always require maximum possible kills with no exceptions unless a monster is literally unreachable, so even pl11 or e1m8 would have to be completed with 100% kills. Second, there could be a "reasonable minimum" category based on some accepted standard of what the minimum kills/secrets should be. This would always be ambiguous and require both community discussion and an authority decision from one of the moderators or veterans, but it would maintain any routes that are too established to make completely invalid such as lv27 extra secret or pl27 extra archvile and would maintain competition on anything that is too stupid to attempt true max like pl11 or close to impossible (?) like e1m8.

But there is such a small amount of "true max" demos (not counting the maps where 100% is easily possible) that a special category is hardly needed. I think we are doing just fine with the occasional "this weirdo actually goes all the way" note on DSDA. If kraflab wants that new category to be recognized, he better start recording a whole bunch of masochistic demos. :)

If a monster nearly impossible to kill (like e1m8), then why don't we just leave it alone? Is it too important to kill those monsters in a situation like su10? What about those maps where 1 or 2 monsters never teleport for some weird reason? Simply skip those monsters, right? :D so I don't think it's important to get 100% kills for those weird maps.

4shockblast said:

Second, there could be a "reasonable minimum" category based on some accepted standard of what the minimum kills/secrets should be.

What for? Nobody is going to bother recording shit like 100% kills pl11. UV-Max is already considered a "reasonable minimum" since ancient c-n times and nobody has needed to add a different category for dumb feats until now.

There is no argument. Anyone can record whatever they want.

Just for teh lulz, somebody should make a "TRUE-MAX-HELL" map that is fun to max and takes about 3 minutes and has about 100 monsters. Then, as a final design element, add a long sequence of super tall and slow lifts that take a total of 1000 hours to ride. Add one zombieman behind all the lifts. 99% kills would take 3 minutes. 100% kills would take 60003 minutes. I'm pretty sure nobody would ever finish "true max" run of that map.

Better yet, let's make a MEGAWAD called "99 OR DIE". All maps would have ingenious designs that make 99% kills easy and fun, but attempting 100% kills would cause the player to die of old age or hemorrhoids.

From wiktionary;

"maximum
ˈmaksɪməm/
noun
1.
the greatest amount, extent, or intensity possible, permitted, or recorded"

So I think looking for 100% kills is simply a quick way of checking whether a demo is 'max'. I think agreeing on what max is on a map by map basis is the best way.

Now to the crux of the matter as to whether we should change the max definition if a new trick is found - I'm inclined to say no. Su10 imp doesn't really add anything to the run and invalidating all the other demos for the sake of 1 extra monster killed is just perverse.

I once had an idea for a map where I would "by mistake" make ALL sectors count as secrets. So uv-max people would have to step on every little piece of detail while hearing DING DING DING. Could be funny.

Aqfaq said:

Better yet, let's make a MEGAWAD called "99 OR DIE". All maps would have ingenious designs that make 99% kills easy and fun, but attempting 100% kills would cause the player to die of old age or hemorrhoids.

Hehe, funny idea :D

Yousuf Anik said:

If a monster nearly impossible to kill (like e1m8), then why don't we just leave it alone? Is it too important to kill those monsters in a situation like su10? What about those maps where 1 or 2 monsters never teleport for some weird reason? Simply skip those monsters, right? :D so I don't think it's important to get 100% kills for those weird maps.

The su10 monsters are far from nearly impossible to kill. I don't see why it wouldn't be important to kill them; if monsters can't teleport at all, then obviously they can't be killed, but there are plenty of cases when runners find convoluted ways to make max or pacifist or whatever category possible.

j4rio said:

What for? Nobody is going to bother recording shit like 100% kills pl11. UV-Max is already considered a "reasonable minimum" since ancient c-n times and nobody has needed to add a different category for dumb feats until now.

I don't see much of a problem in having an extra category. Likely no one will care that much about it, but there are already maxes and nm100s for lv27 and pl27 and even pl11 (Light_Speed's D2ALL TAS max) where the unneeded monsters are killed, so it would make sense to just separate that out to make it clearer what the goals of the runner were.

rehelekretep said:

Now to the crux of the matter as to whether we should change the max definition if a new trick is found - I'm inclined to say no. Su10 imp doesn't really add anything to the run and invalidating all the other demos for the sake of 1 extra monster killed is just perverse.

By that logic, I could record some previously undone map, kill nothing under the pretense of it being "too hard" or "too stupid", and say that every subsequent run is perversely invalidating my run if it kills more things. :V Also, only Bloodite's max missed one extra imp, the previous ones missed more AFAIK, and one of them missed seven, even though only three imps have a particularly hard time missing the player. Would that run be counted as a "max" as well? I believe such exception cases should be kept to an absolute minimum, as they only contribute to increase ambiguity of the rules which is never a good thing. So, as su10 is easy to get a valid max on, all previous runs should be invalidated.

4shockblast said:

So, as su10 is easy to get a valid max on, all previous runs should be invalidated.

Their hard-work achievements are gonna stay in Other category? Hell no ;(

Yousuf Anik said:

Their hard-work achievements are gonna wasted? Hell no ;(

It's a risk you take with almost any run. If someone recorded a pacifist of a map with what seemed at the time like a forced telefrag and someone found a bizarre route that takes twice as long to avoid it, that would end up invalidating the old pacifist. Alternatively, finding some obnoxious trick to beat an old time by a significant margin is not reason to mark that trick as invalid just because it would waste the effort the old runner did on a bad route. Max is already more permissive than most categories in allowing some leeway at all; you don't see any pacifist runners complain that the cyber on e4m6 is "too hard" to kill and should be BFGed instead.

4shockblast said:

so it would make sense to just separate that out to make it clearer what the goals of the runner were

C-n has already judged all cases of their competitive maps one by one to have clear objectives of what is considered a successful max. If there is anything to settle on, it's maps that weren't a part of c-n to begin with. If you feel like a respectable authority, you're free to begin with it. You could start with for example squadron map 7 max rules.

j4rio said:

UV-Max is already considered a "reasonable minimum" since ancient c-n times and nobody has needed to add a different category for dumb feats until now.

Pretty much this. There should be no 'stupidly difficult' monsters or secrets. UV-max has never been 100% kills/secrets in every map.

4shockblast said:

If someone recorded a pacifist of a map with what seemed at the time like a forced telefrag and someone found a bizarre route that takes twice as long to avoid it, that would end up invalidating the old pacifist.

Is this even true? I'm looking at Doom E4M2.

I'm not sure why this is even a discussion. If one demo kills all the monsters and another demo doesn't, then the one that kills all the monsters is a max. If there are annoying monsters to kill then it is a map problem not a demo problem. That is why non-speedrunners shouldn't map.

I think that I agree with Time Of Death's statement and J4rio's statement:

'If one demo kills all the monsters and another demo doesn't, then the one that kills all the monsters is a max.'

'C-n has already judged all cases of their competitive maps one by one to have clear objectives of what is considered a successful max.'

If a person takes the time to max something which meets the following requirements:
1. There is not already a predetermined set of rules set by C-n.
2. There is already a demo made by someone who has not truly achieved 100% kills and 100% secrets.

Then the max the person took a while to complete because of some crazy route change should be honored and the previous demos should be invalidated.


There is an alternative solution I think is reasonable but it sounds to me like it is returning to what C-n already had in place (I will not take part in this however because I am a newer member to the community).

Alternative solution:
If there is a map, that does not have an already predetermined set of rules for max, which causes debate regarding whether or not a secret should be obtained/something should be killed then there must be a person who meets the following criteria determined by longstanding members of the community (more than one person should be decided):
1. Is trusted by the community
2. Is incredibly knowledgeable with the game engine. This means every trick that is possible to pull off in said map (glides, grabs, etc).
3. Is also very experienced with making all sorts of interesting routes

If this person decided by the community is able to prove within a reasonable time period max is possible then the runners of the map should have to truly get 100% kills. Otherwise the trouble monsters/secrets are not taken into account during all future demo attempts.

There is one exception to this alternative solution:
If a map has already been proven to be truly max'd, and that map already doesn't have a predetermined set of rules provided by C-n, then it is required to get 100% kills and 100% secrets.


This is what I was thinking about anyways. I am personally for something along my first idea.

Looper said:

Pretty much this. There should be no 'stupidly difficult' monsters or secrets. UV-max has never been 100% kills/secrets in every map.

Just because something has never been something doesn't mean that there can't be a separate option. For example, I don't see why there couldn't be separate competitions for m1-m7 and m1-m8 runs for Doom 1 given that most people agree that m1-m7 is outdated and makes little sense today other than the fact that it's legacy.

Looper said:

Is this even true? I'm looking at Doom E4M2.

I guess the alternative to a telefrag would be infighting the cyber? Telefrags have always been kind of a gray area in pacifists, so I dunno what the reasoning for that telefrag being counted as pacifist is. At any rate, if someone were to beat my su04 pacifist by picking up the yellow key the normal way and triggering the manc crushers instead of doing the AVJ (or if someone didn't skip any crusher linedefs), I would consider that an invalid pacifist regardless of what their final time is compared to mine.

Killer5 said:

If a person takes the time to max something which meets the following requirements:
1. There is not already a predetermined set of rules set by C-n.
2. There is already a demo made by someone who has not truly achieved 100% kills and 100% secrets.

I'm not a fan of this for a few reasons. One, I don't believe CN rules should be maintained on some sort of pedestal as the sole right way of doing things. Again, back to m1-m7 vs. m1-m8 ep times, not counting a map just for the sake of legacy rules isn't necessarily a good thing, and neither is getting rid of the legacy rule entirely as that simply would not represent the old runs accurately. Therefore, the best solution is maintaining both as viable options to run Doom 1 episodes.

Second, this results in an inconsistent ruling across all mapsets, which isn't a good thing. I'm not aware of the details, but according to dew one of the MM maps had the exception changed by tatsurd because apparently whatever the monsters missed on it weren't actually problematic to kill. If the exception were set in stone, then that change couldn't have happened, and suddenly you have an exception that does not make any sense and most people may not agree with but that has to be kept as is. Across non-CN wads, it would be a similar deal. If I made a map and plopped the Hunted exit room at the end of a 24 minute max and someone happened to get the best possible luck and BFGed all the AVs, I'm sure there would be discussion about whether that would have to be a requirement going forward (not to mention it simply wouldn't make sense to treat the same room differently in two different maps).

Finally, whether you make it an official category or not official (and what even is official in a game where -respawn is an "official" category of all possible options and where categories like Lemmings and Speedfighter and UV Crab exist), someone has the right to run a "true max" category and you can't stop that person from submitting it to DSDA. Would it make sense to just shove it in the existing category even though it absolutely does not aim for the same goals? Even now, I can't say it's a whole lot of fun browsing the DSDA and seeing "-fast UV-Speed with -turbo 173 TAS" above a bunch of actual UV-Speed runs. I believe it makes sense to let people have the option to record whatever they want because regardless of what's a category or not, they can do it anyway.

TimeOfDeath said:

That is why non-speedrunners shouldn't map.

What.

I was half-joking, but these problems wouldn't happen if mappers demoed their own maps.

TimeOfDeath said:

If there are annoying monsters to kill then it is a map problem not a demo probelm.

I agree with this point.

And simply, if the map has very annoying monsters or such secrets or problems, then what's wrong with 99% kills/secrets?

4shockblast said:

Just because something has never been something doesn't mean that there can't be a separate option.

Seems fair as long as the category name isn't UV-max, which I was talking about.