Jump to content
Search In
  • More options...
Find results that contain...
Find results in...
Creaphis

The Two Types of Glides

Recommended Posts

Creaphis said:
If anything deserves a separate category it's Doom runs that use glides, at all.

Perhaps, but isn't already too late for that?

Putting xepop's demo on the top of the podium for e2m6 pacifist runs is hardly any odder.

We got used to "normal" glides, and for me yes, it is much odder. I didn't want to say anything in this topic, it would sound as if I wanted to spoil all the fun you are obviously having with these glitch exploits, for me it is something curious, funny, just once. As myk said, to keep the game interesting, we'd better set the rules now, and I will always stand on the "ultra-traditional" side. Yeah, I am strongly biased against these. What had influence on me were m223 speedruns. I can watch Eugene's traditional runs over and over again, Kristian Ronge's 0:05 run is boring for me.

Share this post


Link to post
Creaphis said:

If anything deserves a separate category it's Doom runs that use glides, at all. Think about it. Much in the same way as in xepop's demo, glides allow you to skip some of the intended play space in a level, and when performing a glide, a player puts his core Doom skills of movement, strategy and reflex on a hanger and just fidgets awkwardly with his mouse instead.

uhuh. check the original glide in map16 and tell me about 'awkward fidgeting'.

I'm not really arguing that demos with glides should be considered as their own category, because that's a difficult line to draw. Keygrabs and linedef-skips also exploit gaps in Doom's programming to skip level content. While we're segregating glide demos, should we also demote the demos that rely on these tricks?

well... the standard glide as you called it in your OP is actually not a glitch or engine exploit. there's a 32unit wide gap, the player goes through it because he's 32units wide too. the glides that skip impassible linedefs or spaces narrower than doomguy's hitbox are possibly exploits on par with grabs and linedef jumps. all of these tricks share one thing though: they happen in-map. :p

Let's just face it: Doom is the sum of all its incredibly glitchy parts. If more records are beaten with void glide demos, cool. Doom running is less about competition than it used to be, anyway, and more about using the vast repository of Doom wads to show cool runs to the world. There's definitely room for void-glide and non-void-glide demos to stand side-by-side in this speedrunning scene.

i'd still count void glide runs as separate, weird oddities rather than comparable speedruns. the analogy to all-ghosts bugs seems appropriate.

Share this post


Link to post
vdgg said:

Perhaps, but isn't already too late for that?We got used to "normal" glides, and for me yes, it is much odder.

Not odder than skipping of two-side walls (see map10)

Share this post


Link to post

looks like it's time for

UV-Void!!!1!

No really, it's a glitch, and those were always used when speedrunning. Besides, what's the chance of getting into void and having a free way to exit switch? Lots of maps utilize ground being a trigger of level finish. Also huge maps with all those rooms barricade player to get the trick thing done. I doubt it will be so heavily usable in most cases anyways.

Share this post


Link to post

No, I don't see any need for a new category in general. There are already a great many tricks routinely used by speedrunners to break the intended gameplay of a map. No doubt when each was discovered, there were some misgivings about them because of what was therefore rendered obsolete, but they have all been accepted as part of speedrunning. Outlawing new tricks, or pushing them into some special category, would run against the history of speedrunning and make the whole genre stagnate.

Personally, I am delighted to see Doom speedrunning constantly moving forward, and players finding new ways to exploit quirks. I also feel that any quality demo will always be welcomed by the Doom community. The runner simply needs to describe what rules they have followed (e.g. "UV Speed without tricks" or "UV Speed without entering the Void", "NM 100% Secrets moving only backwards and only using the rocket launcher", etc.).

The compet-n ethos has always been "if the engine allows it, it's fair game".

There could be an argument for a special category "without tricks" (but that's potentially messy because of what you define as a trick), and this has been used for certain maps or a more or less formal basis (remember Sedlo's runs on lv24, for instance), and for the Max category there have been such distinctions on some maps where a true max is problematic. Quake speedrunning also has a strong tradition of such distinctions.

But the raw traditional categories should not be altered to disallow new ideas. I don't think we even have the right to do that, given that there is such a rich history of demos and categories that predate this forum and most of its users.

Also, as j4rio and others have pointed out, there probably aren't many maps where this does have an impact on speedrunning. Firstly you need a set-up that allows the player into the void. Even if it is every time this specific geometry occurs, that's not so many (on a quick look through Doom2.wad, I found only map01 and map30, and in neither case did it look a viable route). And then you need a way to reach the exit, which will not often be in a place accessible from the void. And often it is not a switch at all, in which case the player will need a way to break back into the map nearer the exit. Overall, I think this type of trick will have far less widespread impact than such discoveries as 32-unit glides or suicide exits.

myk said:

If that kind of play were standard, then why aren't we waiting for lucky intercepts glitches to allow us to run quickly to the exit?

Because it is very very rare for them to be helpful. You need an "exit if health <10%" sector, which is a very rare feature in maps, and you need to be able to reach it, which is far from guaranteed. It's an awesome trick, but only used in one map. In other cases it just trashes your attempt.

myk said:

You would say it's valid, but you also felt sr50 on turns was valid, back when you made your TAS movie, with the "if it's technically possible, it's allowed" principle.

Andrey based that on the idea that this was possible using a joystick (as well as keyboard and mouse) and a third hand. He has since stated that he was in error and allowing sr50 on turns was a mistake.

myk said:

I was thinking, anyway... was this so unexpected? What led Gusta to believe hitting the exit switch from outside was not possible, with the E3M6 precedent and all?

That's not a precedent, because that switch can be pressed through the wall. This one can't - the wall absorbs keypresses. You need to press at the actual point of interception. That's why people gave this up as impossible - they tried pressing through the wall as normal, and failed.

Discussion split off from glides thread to avoid derailing the discussion of the mechanics of this new trick idea.

Share this post


Link to post

Yes, I meant that it didn't look like a viable route to exit the map, short of some additional new trick.

But the conclusion so far seems to be that this geometry allows the player through fairly reliably (right angle at 45 degrees to the grid, with the vertex at an intersection of 32-unit gridlines allows the player through in a west-to-east direction).

Share this post


Link to post
entryway said:

Not odder than skipping of two-side walls (see map10)

In e2m6 it's gliding to the void. In map10, pl2 map07 they glide to other sectors. It's Different

To the people who are against these void glides for the normal categories but agree about map10 and pl2 map07 type.
In e2m6, if the blue key wall was glued to the exit room. Would you accept that trick? He would glide directly inside the exit room.
In that case, it should be accepted the same way map10 and pl2 map07 were accepted, unless you want a transparent wall condition too :P

I'm not yet sure about accepting this void glide trick just because I'm not sure which maps it could break. If I understand it right, could it break Map23 for example? You can glide to void and then into the exit

Grazza said:

Even if it is every time this specific geometry occurs

Which geometry does it need? Is it any 2 intersecting impassable linedefs so you can wobble them? And which height does the player has when he's in void? I was gonna say he retains the height from the sector he just left but entryway's demo on map30 confused me.

Grazza said:

Because it is very very rare for them to be helpful. You need an "exit if health <10%" sector, which is a very rare feature in maps, and you need to be able to reach it, which is far from guaranteed. It's an awesome trick, but only used in one map. In other cases it just trashes your attempt.

Should absolutely be another category if the player uses intercept overflows to exit these sector types for obvious reasons

Share this post


Link to post

I think they should count as standard UV speed - if the Doom engine allows it, it's fair game. Also like Grazza i'm spectical that this'll turn out to save time on more than a handful of levels.

Share this post


Link to post
kimo_xvirus said:

I'm not yet sure about accepting this void glide trick just because I'm not sure which maps it could break.

I think very few, most likely, unless it proves possible to break through a wider class of geometries than has so far been demonstrated. And even then, there may not be many maps where it may be useful in terms of allowing the player to exit faster than is possible by other means.

If I understand it right, could it break Map23 for example? You can glide to void and then into the exit

I don't see how, unless you found a way to break out of the map and then break back into it near the exit.

Which geometry does it need? Is it any 2 intersecting impassable linedefs so you can wobble them?

This is still unclear, but the only proven instances so far feature a right angle at 45 degrees to the grid, with the vertex at an intersection of 32-unit gridlines, allowing the player through in a west-to-east direction. Discussion in this thread.

And which height does the player has when he's in void? I was gonna say he retains the height from the sector he just left but entryway's demo on map30 confused me.

Movement in the void is a lot more complicated (and weird) than most people realize, as there are height changes related to sectors of the map itself. It's absolutely not the same as noclipping. A typical result of breaking out of a map will actually be getting trapped in the void, unable to go anywhere useful. To experiment with this, you can use IDCLIP to move into the void and then IDCLIP to turn clipping back on. When doing my zombie/ghost demos, I found that after breaking out of Doom2 map04, the only way to reach the exit was by going all the way through the void and wrapping round to the other side - and that was despite being able to walk through walls!

Demo referred to attached.

stuff208.zip

Share this post


Link to post

It seems premature to talk of banning void glides. Wait until it starts obviously ruining the game before considering it. But as has been mentioned, speedrunning is pretty well "ruined" already with all the unrewarding barely-doable tricks. It's the purists who ought to start a new category - "UV speed, 100% secrets, 100% keys" for instance.

As for the rarity of these things, looking through the original wads (only the maps that looked exitable) I found void glides on E1M3, E1M6, E2M1, MAP30 and MAP31, but with none of these was I even able to reach the exit let alone press the switch. But it's possible that in a particular megawad the right geometry might occur very often.

Meanwhile, here's a second void exit demo. Probably useless, but it should please the trickaholic crowd. E2M5 normal exit nomonsters (TAS) - E2M5x036:

e2m5x036.zip

Share this post


Link to post

Creaphis said:
Doom running is less about competition than it used to be, anyway, and more about using the vast repository of Doom wads to show cool runs to the world.

We aren't talking about DOOM running in general here, nor of deleting such demos from the forums or something.

Grazza said:
There are already a great many tricks routinely used by speedrunners to break the intended gameplay of a map.

Making a key grab or sr50 invalid, for example, is extremely impractical because these occur in many, often minor or irrelevant situations even if not attempted on purpose. Other hacks, such as this one, have a much greater effect and are easier to manage.

Outlawing new tricks, or pushing them into some special category, would run against the history of speedrunning and make the whole genre stagnate.

That's not a response to what's being proposed. The possibility to escape into the void, when available, makes any run through most of the map pointless. Regular glides, arch-vile jumps and key grabs already do this to a point but still require you to stay within the map.

As exemplified by Quasar and many other situations in DOOM or the SDA, possibilities have arbitrarily been disallowed in order to make running more varied and interesting. This is not "a ban" but rather that they require an admin's determination to be valid, since they are new and change the nature of the game (now you can run around in the unmapped area.) In another recent thread about "solo net" demos, a rule about not respawning when killed is mentioned. Respawning is a technical possibility and it is not allowed at Compet-n.

Personally, I am delighted to see Doom speedrunning constantly moving forward, and players finding new ways to exploit quirks.

I think it's a cool find, but it doesn't really make things "move forward." But you dedicate almost all your DOOM-related time to glide demos :p

The compet-n ethos has always been "if the engine allows it, it's fair game".

If that were the case, it would be a TAS competition.

But the raw traditional categories should not be altered to disallow new ideas. I don't think we even have the right to do that, given that there is such a rich history of demos and categories that predate this forum and most of its users.

Whenever an oddity came up, Adam Hegyi or whoever was the admin would check it up, consult with other runners and take a determination about it. You're right about us not being in any position to do that in this case, as Compet-n is effectively "frozen."

Overall, I think this type of trick will have far less widespread impact than such discoveries as 32-unit glides or suicide exits.

The impact is in what it does to the game. It alters what the map is, in practical terms. It's not a trick that is being questioned (proposed as invalid,) since I didn't propose that glides be treated separately, but the fact that the player is off the level area. That is the novelty.

Andrey based that on the idea that this was possible using a joystick (as well as keyboard and mouse) and a third hand. He has since stated that he was in error and allowing sr50 on turns was a mistake.

Right, these ideas can be argued against and people convinced they weren't right in the first place.

That's not a precedent, because that switch can be pressed through the wall. This one can't - the wall absorbs keypresses. You need to press at the actual point of interception. That's why people gave this up as impossible - they tried pressing through the wall as normal, and failed.

Differences aside, it's very similar, and it's not too hard to hit the exit switch in E2M6 from the outside. Not much harder than in E3M6, at least.

Discussion split off from glides thread to avoid derailing the discussion of the mechanics of this new trick idea.

I don't think it derails the mechanics discussion. If anything, this discussion is thrown off context if split. Discussing validity without examining technicalities is awkward, and technicalities come up due to discussing validity (see kimo's height question as an example). With that, merged back.

kimo_xvirus said:
In e2m6, if the blue key wall was glued to the exit room. Would you accept that trick? He would glide directly inside the exit room.
In that case, it should be accepted the same way map10 and pl2 map07 were accepted, unless you want a transparent wall condition too :P

That's right. Adam Hegyi already validated such a possibility when he said OK to the level 10 trick (I think that was the first instance.)

cack_handed said:
Wait until it starts obviously ruining the game before considering it.

Not that we can do anything in the first place other than somehow end up agreeing, as I explained, but retroactively trying to make some trick invalid is extremely silly. You can expect people to use it "for the time being" and then suddenly make a bunch of demos invalid. Also, that it is rather useless may be another reason to make it invalid. Rather than have players waste their time testing walls all over the contours of a map, just remove the possibility so they can concentrate on the proper level.

It's the purists who ought to start a new category - "UV speed, 100% secrets, 100% keys" for instance.

It's not a matter of purism. This is about Compet-n. "Anything the engine allows is valid" is also a form of purism, after all. If intended-route purism were desired, it'd have to be done in another space. These demos aren't going to disappear even if we agreed on whether they are valid or not, anyway. You can do them anyway and they still have their value. If we disallow them, what we gain is to still encourage "normal" runs in levels where this exploit exists. If you allow them, and find a map has this near the beginning, with an accessible exit, why even speed that level normally anymore?

Inventing "purist" categories for Compet-n is much more extreme and unthinkable than concluding that going out of the proper map bounds is not valid for Compet-n. It had never come up before, so the admin (Adam Hegyi) never had a need to address it.

Share this post


Link to post

Looks like a cool exploitable bug for me. I'll use it, since i dont play for compet-n records anyway. Some maps would of course become more boring, but well, shortcuts are shorcuts.

Would be cool to clarify compet-n rules about that stuff, but as i understand, no one can do that besides AdamH and there will always be silent controversy about the legitimacy of void-demos. We gotta do something about it :]

By the way myk, let's suppose there'll be some neat noclip-trick which will allow to move anywhere in the level. Would it be legit to go through walls inside the level boundary in your opinion? And what about leaving the level boundary with the same trick? I still not quite understand your point - is it about "modifying" the level or more about the boundaries themselves. Like, what about the good old doom2 map24 shortcut? You're practically walking on a ledge which is beyond the level boundary at that moment, is it ok to use the things that arent inside the main box?

Share this post


Link to post

jongo said:
We gotta do something about it :]

We've been doing pretty well with things as they are. To "do something" we'd need to get word from Adam and produce conditions that make competitive play more abuse-proof.

By the way myk, let's suppose there'll be some neat noclip-trick which will allow to move anywhere in the level. Would it be legit to go through walls inside the level boundary in your opinion?

If you disallow no-clip bugs, it should be when they happen, and not dependent on what the player does, as that would make things complicated. The standing of all-ghosts demos in Compet-n is not clear.

Like, what about the good old doom2 map24 shortcut? You're practically walking on a ledge which is beyond the level boundary at that moment, is it ok to use the things that arent inside the main box?

You mean the ledge jump at the end? By "out of map bounds" I mean clearly out of the sectors that comprise the map. I don't see that happening there, especially in any way like in the void demos.

Share this post


Link to post

myk said:
Inventing "purist" categories for Compet-n is much more extreme and unthinkable than concluding that going out of the proper map bounds is not valid for Compet-n. It had never come up before, so the admin (Adam Hegyi) never had a need to address it. [/B]

Or we can just not invent extra purist categories. Also I'm not following the argument that since Compet-n rules don't specifically address the question of whether voidspace demos are valid means they default to not valid. Quite a lot of tricks haven't been officially designated as valid. On the other hand the rules are quite explicit about what is required for a particular demo to be a valid entry. Which engine version, which exits on which map, what conditions must be met for it to be a pacifist or respawn demo, how many secrets required, and so on. The demos posted so far meet all the requirements to be valid UV speed or pacifist, therefore they are such.

This is reminding me of some early Compet-n history. One of the very earliest game changer speedrunning tricks to be found was the e2m2 exit switch, by Steffen Winterfeldt. Because the Doom engine doesn't check the relative height of the player to the switch he's trying to hit, you could run straight to the exit, completely ignoring the "designed" route through the level. At around the same time the strafejump to the exit was also discovered. So for a brief period runs were actually being accepted to Compet-n on all three routes - the standard no-trick route, the switch trick route, the strafejump route. The first switch trick demo was by Simon Widlake, who called it "cheating" his way through the level. And he was still accepting submissions on the old route more than ten months later.

Obviously there was some doubt in his mind as to whether "tricking" one's way through a level really counted as a proper speedrun. But the switch trick became the accepted route, and only submissions beating the switch trick time would now be accepted. That was really the point at which the floodgates opened on what kind of bug-exploiting tricks would be permissible. Key grabs, archvile jumps, wallrunning, strafe50, suicide exits, glides, all the variety of modern Doom speedrunning flowed on from that decision. Now suddenly making a stand fourteen years later and declaring a certain kind of movement bug verboten in Compet-n seems like an odd stand to take.

Another thought: what about keygrabs where the Doomguy is reaching through the voidspace, like the BFG grab in e4m2? Arguably these take him outside the map - we might need to place a little asterisk beside all those records.

Share this post


Link to post

I'm quite surprised that this trick is even worth a discussion, considering there is but 1 demo showing a way to make getting to map finish faster. I'd be actually really entertained if someone actually makes up some of these "speedruns" of iwads/famous pwads consisting of getting out of map and somehow finishing level.

Share this post


Link to post
j4rio said:

I'm quite surprised that this trick is even worth a discussion, considering there is but 1 demo showing a way to make getting to map finish faster. I'd be actually really entertained if someone actually makes up some of these "speedruns" of iwads/famous pwads consisting of getting out of map and somehow finishing level.

If some levels can be exited this way, it stands to reason that many levels can be exited this way, because DOOM is a very self-consistent game engine :)

Share this post


Link to post

Ryback said:
Also I'm not following the argument that since Compet-n rules don't specifically address the question of whether voidspace demos are valid means they default to not valid.

I don't think that argument was made. They simply don't address the issue, mainly because apparently it hasn't been an issue till now. Even literally speaking, though, what the rules say does depend on what you consider to be "the level." Is the void part of the level? The meaning of level will depend on one's position, especially if no one cared about what it meant in this sense or specifically before the issue.

I do wonder what was said (particularly by Adam) back when Gusta was helped off E2M6 by that demon.

Now suddenly making a stand fourteen years later and declaring a certain kind of movement bug verboten in Compet-n seems like an odd stand to take.

The movement bug would be the glide, which alone is not being questioned. If everything were so clear from interpreting the rules literally and from making generalizations based on existing historical precedents, there would be no need for a Compet-n admin. Aside from my preference, my main point is that this is something new and weird that would call for an admin's decision, so it's best to put a parenthesis around it. Doing what Widlake did early on for the E2M2 tricks, allowing different demos, seems reasonable here.

Another thought: what about keygrabs where the Doomguy is reaching through the voidspace, like the BFG grab in e4m2? Arguably these take him outside the map - we might need to place a little asterisk beside all those records.

I wouldn't seriously argue that the player went into the void in such a case.

Share this post


Link to post

It's just a trick, resulting in a faster UV Speed demo.

I saw a HL1 speedrun where most outdoor maps are beaten this way.

Share this post


Link to post

This could always have a debate like killing lost souls in UV Max. The problem is the highest authority (Adam) didn't address if doomguy being completely in the void is allowed in the standard Compet-n categories. But in the end, it's all up to people choice how they record their demos.

In the tas speedruns community, some people were talking about the game breaking glitches. Someone said if there was one glitch at the start of a game that gets you to the ending of a certain game and call it a speedrun it would be lame. This is the idea, Some glitches are lamer than others.

Likewise, I'm pretty sure if someone were to record a demo with a memory glitch that lets you exit from Map01 to to Map30 in NM. Everyone would prefer Henning's 30nm more. That theoretical memory glitch demo though would still have its technical value in that it completes the game on NM faster, both record demos can be equal in some sense :P

Imo, this trick/glitch is the lamest we had in doom till now. Doomguy being in the void in a place where the author could've never imagined the player be. It doesn't feel right to me to let xepop's e2m6 demo completely overtake Xit's UV Speed and AdamW UV Pacifist. I judge this based on the trick itself, not because it was only done in few maps. No offense to xepop's demo, it always had its value.

If I would update the Compet-n tables, I'd put xepop's demo just aside Xit's demo in the UV Speed table and aside AdamW's demo in the UV Pacifist table and organise the tables in some clean way to distinct the trick. Same thing done to this trick done in any other map. And if someone were to record an E2 movie with this trick, it'd be the same.

Compet-N is dead, if the community would distinct the void demos here and at DSDA. We may take into consideration doing the same for all of the other tricks. After all, if we had the skill/time/will we would optimise every map in every possible demo category.

Share this post


Link to post

Vile said 3 years ago:
Eventually we'll find engine bugs that let the Doom guy break down any wall he wishes and speedruns will become obsolete.

http://www.doomworld.com/vb/showthread.php?s=&postid=695872#post695872

kimo_xvirus said:
If I would update the Compet-n tables, I'd put xepop's demo just aside Xit's demo in the UV Speed table and aside AdamW's demo in the UV Pacifist table and organise the tables in some clean way to distinct the trick. Same thing done to this trick done in any other map. And if someone were to record an E2 movie with this trick, it'd be the same.

I was thinking about something similar, I agree with this.

Compet-N is dead, if the community would distinct the void demos here and at DSDA.

Pardon, I don't understand?

Share this post


Link to post

myk said:
what the rules say does depend on what you consider to be "the level." Is the void part of the level?

It's not entirely separate, as what happens in the void can affect the rest of the level. eg PEs spawning lost souls in the void, whose movement can be heard and will have an effect on the RNG. Or Archvile attacks where most of the blast is absorbed by the void, leaving the player almost untouched. Ghost monsters can go wandering in the void, and I think there are levels where they have to cross voidspace before showing up to be killed by the player. So I would lean toward considering the void just another part of the level.

Share this post


Link to post
Ryback said:

It's not entirely separate, as what happens in the void can affect the rest of the level. eg PEs spawning lost souls in the void, whose movement can be heard and will have an effect on the RNG. Or Archvile attacks where most of the blast is absorbed by the void, leaving the player almost untouched. Ghost monsters can go wandering in the void, and I think there are levels where they have to cross voidspace before showing up to be killed by the player. So I would lean toward considering the void just another part of the level.

Perhaps even more telling is that there is no function in DOOM which can test a point and tell you if it is in "valid" space or not. There is simply no distinction.

The most important point-testing function, R_PointInSubsector, will return a valid subsector regardless of the player's position, which is why void areas still display the properties of a single valid subsector at a time. This is because the BSP tree provides a complete mapping of space - there is no point "outside" of it.

Thus, there is no real mathematical distinction between valid space and void space, within the engine***. That means the only distinction to be made is one which is philosophical.

Although it is possible to move outside the blockmap ;) Objects which do so will become no-clipping with respect to each other. So I suppose you could define the REAL void as the space outside the blockmap bounding box. The only problem with that definition is that it excludes spaces within the box which are essentially identical aside from still having object vs object clipping :P

*** One *can* be made, by adding an in-out polygon decision algorithm and using it on completed subsectors - but such is an extensive and highly technical extension that must be made to the engine first ;)

Share this post


Link to post

There shouldn't be any "lameness" factor deciding whether gliding to void space is any different than other engine quirks. I've yet to hear a good understandable explanation why this should be treated differently.

This map just happens to have perfectly structured void space so one can run to the exit like nothing. It doesn't have to be and/or usually isn't like that. It's possible one might have to use rocket jumping, glides, whatever, for navigating in there. Would be interesting. Currently it's affecting 2 known maps e2m6 and icarus map29. I wonder why in icarus it's a harder to do. Is there something else affecting it? Doom2 map30 is the definitely the easiest one.

Share this post


Link to post
Ryback said:

It's not entirely separate, as what happens in the void can affect the rest of the level. eg PEs spawning lost souls in the void, whose movement can be heard and will have an effect on the RNG. Or Archvile attacks where most of the blast is absorbed by the void, leaving the player almost untouched. Ghost monsters can go wandering in the void, and I think there are levels where they have to cross voidspace before showing up to be killed by the player. So I would lean toward considering the void just another part of the level.

Now that doomguy can enter the void, maybe the max definition should be updated to say that lost souls must be killed. ;)

Share this post


Link to post
xepop said:

There shouldn't be any "lameness" factor deciding whether gliding to void space is any different than other engine quirks. I've yet to hear a good understandable explanation why this should be treated differently.

I thought about it again, I'm still not sure yet about accepting it or not :/ This can break too many maps. I noticed in cack's hangar13 run, he glided through a right angled intersection of 90 and 0 degrees walls. In e2m5 and e2m6 it's 45 and -45. In map10 and pl2 map07, it's not a right angled intersection. The only thing they have in common is intersecting impassable linedefs that you can wobble. For this reason. I think there is a huge amount of possible skip glides.

Share this post


Link to post

Presumably you'd need to reach a point with the same geometry (or rather the inverse of it) and break through in the same way. Or if other geometries that allow this trick are discovered, then the inverse of one of them.

That is, if you can't press an exit switch from the void, then you need places to break in and break out that are useful for reaching the exit, and to do so faster than on a normal route. And of course you need a path via the void between those two points, and often there won't be such a path.

That's one major reason I'm maintaining that this won't affect all that many maps, and that some people are getting overexcited about this trick destroying speedrunning or whatever. To me it is a cool (and absolutely not lame) new trick idea of the type speedrunning seems to encounter every couple of years. A few maps will have a new speed route as a result, and some creative route ideas will be devised that include this theme.

kimo: cack_handed's hangar13 run didn't feature moving into the void. It's actually very similar to the e1m3 glide, where there is also a right angle at the point of intersection. So actually this idea is about a decade old. :) The main idea from that demo and its analysis is that it is an obstruction to your movement away from the two-sided linedef that is necessary for this type of glide. That obstruction can come from a sloping wall (as in most of the demos where this trick is used), but it doesn't have to. I guess it could come from a coop partner, in which case pl15 looks ripe for this.

Oh OK, I understand: you're viewing this as identical to glides into the void. Until someone shows that you can reach the void via a straight section of wall, that's conjecture. Two-sided linedefs clearly have some different properties compared to normal walls (as seen from the wallrunning and grab possibilities), so I don't think one can assume that they behave in the same way for glides too. I think that if they did, then players would have been falling out of maps on a regular basis these last 17 years.

Regarding lameness, there used to be those who considered any use of strafe-50 to be lame. That point of view seems incomprehensible to most of us nowadays, doesn't it?

Share this post


Link to post
Grazza said:

That's one major reason I'm maintaining that this won't affect all that many maps

Right

Oh OK, I understand: you're viewing this as identical to glides into the void. Until someone shows that you can reach the void via a straight section of wall, that's conjecture.

Looks like they aren't identical. I tried putting void space next to the map10 fence glide and tried gliding through but I didn't succeed in so. I tried the same with lots of right angled walls but didn't succeed too. Maybe others can succeed in them if they get lucky but I don't hold much hope.

All the void glides so far have been in 45 and -45 angled walls. even if you flip them horizontally to be 135 and 225. They don't seem to work.
Map30 has both examples at the beginning, and the right side is very easy to glide through probably due to the hanging body. I tried the left side for a while but didn't succeed.

I think due to this and the need of a good void path, it doesn't look like this trick can break many existing maps at all. Man I changed my mind, I think this trick should be treated like the other doom tricks now :)

Spleen said:

Has anyone figured out how to exit the void yet? As in, going back into the normal part of the level.


Yes, the same way you glide to the void. I made a map/demo showing this. The glides are in the same shape as the map30 one to make it easy for me :)

voidinout.zip

Share this post


Link to post

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×