On A Rail Uncut - Re-Adding Cut Areas/Scenes (Expansion Project)

@Maxwell, I’ll address your schematic a bit later, when I have a bit more spare time.

I decided to take a look to see what would be the best method of addressing the spatial issue, and if indeed it was possible/feasible. Long story short, I can address MOST of the spatial concerns between the maps with a little bit of magical jiggery-pokery, but addressing ALL of them is a little out of the question.

To help me visualize the problem, I actually constructed a map composite in Hammer using the basic map brushwork. While Wangman’s mockup was nice, because I’m a mapper it helps me to visualize things better through a medium I can easily relate to and understand.

This schematic represents the current spatial layout of my maps, when I’ve made NO modifications to any of the Black Mesa maps:

As you can see, there’s nowhere my maps can fit into this layout which doesn’t make the traveling feel pointless, an issue which has already been brought up. Mapping B1 into the empty space on the left seems like a neat idea but ultimately creates the problem of pointless travel, you’d be leaving B at one point and coming back to that same point, just a bit higher up, to transition to C a lot later, and that ain’t good.

The only simple way to resolve the spatial issue using this layout, would be to make my B1 map immediately come after A1, and have it predominantly travel to the right. This is a viable option as it resolves a few issues, but I ultimately don’t like it as it disrupts the chronology of Half-Life 1, and that’s not something I’m particularly considering doing. However it DOES resolve the the spatial issue and the separation of the launch/priming which would be present otherwise, so this IS a viable option, but not one I’m considering currently.

As predicted, Wangman’s solution of rotating the C map 180 degrees has proven to be the most viable option, at least in my mind. Having looked closely at the problem now, I’ve also found a pretty simple way to make it work well without much input from me - a pretty elegant solution if I do say so myself.

Rather than doing as Wangman suggested and rotating the rocket silo/entire C map, and the silo on the B map 180 degrees, which would be an editing nightmare (and DEFINITELY not something I’d be willing to do), there’s actually a much simpler way. The only part of the C map which the player ever observes from the B map is the skybox (representing C) above the silo. By rotating THIS 180 degrees on the B map I can make it appear as if the C map has been rotated 180 degrees. That is definitely not difficult to do. The player will be none the wiser. The only part of the B map which the player ever directly sees from the C map, is the innards of the silo itself. Rotating THESE innards 180 degrees is also not an issue, as, unlike on the B map, they aren’t actually connected to anything on the C map. By rotating these 2 opposing elements 180 degrees, it will appear that both maps have been rotated 180 degrees.

This solution is generally pretty brilliant. It avoids any HEFTY editing which is very time consuming, difficult, and could cause any number of problems. It creates a layout which does convey a sense of travel and progression, while also making spatial sense.

This layout does have one MAJOR issue though, one which Wangman didn’t consider. Though I forgot to circle it on the schematic, take a look at the transition to Apprehension, which on the rotated schematic is the very visible green curve sticking out the center of the C map at the top. It overlaps the B map by a fairly hefty margin, and, sadly, they’re on the same vertical level.

The overlap is visible here:

I added Apprehension’s first map joined together with the C map. I then selected it so it’ll be in red. The overlap is where the red meets the brown. As you can see this is a pretty damn serious issue too and I’m not too sure what can be done about it.

I’ll sit and think on it. I have to go and do some uni work right now, so I’ll get back to you later. Hope this clears things up a bit though.

Perhaps making the apprehension tunnel have an incline (or decline, depending on which you want) and moving the whole thing up or down a number of units?

Your solution doesn’t make anything easier compared to what I said previously. To do a spatially correct remake you STILL would need the following information from BMS B and C maps:

  1. the segment from the rocket to the end of B
  2. the segment from the beginning of C to the rocket
  3. the vertical distance between these two segments
    Again, these segments and their correct spatial orientation would need to be held in place and the B1 map constructed AROUND THEM to resolve all issues.

What you’ve basically done is address the chirality problem w/the moon and rocket platform at the B map instead of the C map. The one advantage that this does offer is that it allows you to rotate the moon and rocket platform AT ANY ANGLE. This negates the third spatial artifact that I mentioned in my diagram which was the scaffold BELOW the rocket platform, which has a rectangular symmetry. Since you’re editing B, you can rotate this along w/the rocket and moon.

So, to solve your problem w/Apprehension, look at your second map and rotate C 90 degrees clockwise relative to B. Problem solved. Now you will just need to rotate the moon and the entire rocket apparatus 90 degrees COUNTERCLOCKWISE in B to keep the spatial orientation correct. Assuming the rocket platform and supporting scaffold is shaped as a square rather than a rectangle, this should be easy to do. Note that the arc that you can use to design B1 is a little narrower with this new solution in theory, but this problem is pretty trivial. You could still use my E redesign schematic from the pitch bible to design your section. Just mirror the turns or angle them slightly as needed.

Also did you rotate the A map differently to accommodate your A1? How you orient it should be irrelevant but its for my own edification.

Shooting down some BS
@Kweebo
There is no problem whatsoever w/the TOW launcher. It is an antitank, antivehicle weapon that is placed in OaR to destroy trams which makes sense. My E redesign schematic shows how the one-shot launcher in BMS could be used w/o any problems.

@Max
What exactly is the HL correlate of your proposal? Aside from an obvious lift area, where would this go? Don’t you think there’s too much content packed into this sequence?

Wow, thank you! :slight_smile:

Also, I don’t know much about Hammer, not because I’m lazy, but because I’ve never been able to get it to work on my computer. But could an effect similar to rotating a map be produced by rotating the entrance of a map, and adding a turn onto the map itself?

I’m thinking it could be placed near the beginning of the first post-rocket-lift map, thereby keeping the player away from the surface a little longer. It doesn’t have a precise HL correlate, but then again, neither do a lot of things in Black Mesa, particularly in On A Rail. It’s an amalgam of several parts of the original On A Rail (the barnacle tram lift, the explosive elevator puzzle, fighting marines on a tram lift)- and since, even with this expansion, BM OAR will be shorter than the original, including complex, vivid segments incorporating multiple segments from the original will be essential in restoring the original’s content. I don’t really think there’s too much content in this sequence. It follows a sensible order of events, with simple puzzle-solving followed by a quick action bit, and the area makes sense (an HECU-occupied tram lift).

Seems like adding some train turntables/elevators to show that these maps (end of “A1” and “B”) are on different levels, may fix the problems. Or you could just tell yourself that “B” is below the first map of Apprehension. Because when you come up from B to C the journey up the elevator is farely large, I think 3 floors off the top of my head. But, when you travel from C to apprehension you only go down 1 floor. If you think of it like that it makes this problem irrelevant.

Personally I just don’t care. OaR is such a maze with going down then up then up then down and down again while taking all kinds of turns that I seriously don’t think most players would notice architectural impossibilities between the map they are on and one they played earlier. If it’s not something that’s noticeable without making a schematic of all of the map layouts like that then I just don’t see it as much of an issue honestly or maybe I’m just missing the point here. Anyways that’s just me. I know you guys will come up with some trickery to fix this.

In a way it could actually be kind of cool in a subtle mind fuck sort of way that disorients the player without them really noticing what it is that’s giving them that feeling. This would be akin to how the set designs of The Shining worked.
https://www.youtube.com/watch?v=0sUIxXCCFWw

It makes EVERYTHING easier. You don’t know what you’re talking about in this particular field relative to the editing aspect. It is so much easier to rotate elements which aren’t connected to anything, such as B’s skybox and C’s silo that it would be ridiculous for me to try and do it any other way. Rotating the ENTIRE C map is an absolute nightmare, as is rotating any entire map. In fact, just making modifications to the maps in general is something I’d very much like to avoid.

This solution works fine, unless someone can prove otherwise. The only frames of reference you EVER have between B and C, to figure out their spatial orientation, is seeing C’s skybox in B’s silo, and seeing B’s silo on C. The maps don’t need to be ACTUALLY reoriented because I can adjust their RELATIVE orientation through these 2 elements - as they are the ONLY reference points… It’s a FAR more elegant solution than any other approach, addressing the problem without creating any new ones.

Neither 1) or 2) of your points are particularly important in this regards, because I’m going to disconnect them from one another anyway, that’s the objective of this “reorientation”. That’s not really a problem, unless you’ve seen something I’ve missed, which I’m pretty sure I haven’t.

The vertical space isn’t an issue either, provided I make B and C stay on the same level as one another - which is really simple. All I have to do is keep vertical travel on B1 net neutral and that works out just dandy.

The ONLY problem I can readily see with this setup is the Apprehension overlap, and I’m pretty sure it’s not…too large an issue. Unlike the rest of B and C the players will have nothing to reference its position against. I can probably do a little bit of magical jiggery pokery to make that work fine.

As far as I’m concerned this seems to be BY FAR the optimum solution. I understand Wangman you want everything to be perfect but you have to understand I have to achieve a balance between amount of work and the results. Rotating entire maps is A FUCKLOAD of work, for a result which will probably be equal to my rotation idea, possibly even worse - a LOT of things can get mucked up in the process.

If you see a big flaw in my plan, and you think I’ve missed it, and you still don’t agree with me - feel free to try and explain it to me - but you’ll have to try and make it clear. There’s every chance I’m missing something here and this type of spatial orientation has always been a weakness of mine. Having said that however, I am fairly confident that I’ve nailed this problem, hard.

As for your last rotation question, I slotted in all the other OaR elements relative to A1’s default alignment in the editor. As I said, I was trying to create a layout map that I could easily relate to, and as I’m so familiar with the top down from A1, that seemed the best way to do it. But, A’s default alignment/rotation matches up with A1’s default alignment. It’s B and C where things start getting funkay.

Not possible to end the map and then load the next map, so u don’t need to rotate it, but the player ‘warps’ during loading from one point to another?

So I don’t think they need to be lined up perfectly, just give us the impression they are lined up

I was under the impression you were not going to edit existing maps whatsoever because of issues related to decompiling / recompiling, and that this was the root of the dilemma. It seems to me that, as long as you are editing them, the whole issue could be averted by deleting the exit and entrance tunnels (of B and C respectively) and replacing them with ones that go in different directions. Or would that somehow be equally problematic as rotating the entire thing?

Or here’s a more conservative idea, which I think would give you a lot of freedom while keeping edits to existing content to a minimum: Place your blockage directly in C2A2B’s exit tunnel. Then, swap the inaccessible door on the side of this tunnel (next to the health and suit chargers) with an accessible door which opens into your maps. The player will need to leave the tram behind and switch to a new one at the beginning of your map, which can be anywhere and in any direction you choose.

I was initially VERY against decompiling the maps but as time has progressed it appears it’s going to be pretty much unavoidable, which is a shame.

Editing the tunnels is not a viable option because the maps are so closely oriented and are tied together. Pretty much no matter how I edit the track at the end of B (shown on the schematic), any travel between that tunnel and the C opening (even if I move around the C opening) would be quite obviously a serious loop.

No matter how you alter the two paths it still creates the problem of the area between being extremely small. In order to compensate for this smallness I’d have to make the path into a hairpin, which would be awful.

The only way that would really work is if the circled area, on B, was moved to be adjacent to the silo (the big circular structure) - so B’s end tunnel curves 180 back towards the left of the silo. I would then have to alter C’s tunnel so that it was 90 degrees oriented so it would be pretty much on the very right of the schematic. This would still require extensive map editing and re-orientation anyway, and is pretty much the only way to make your solution work well. I could also reverse it (swap B to the right and leave C alone), which could work alright, but again, there’s a serious space issue involved there.

EDIT: Figure I’d illustrate what I was talking about:

This would still require me to edit both B and C, but it IS potentially the second best solution which I can see. Still, unless someone can point out to me why rotating the skybox on B, and the silo on C, wouldn’t work as well, I think I’ll stick with that. This can be my plan B. It has a pretty big space problem though, and it’s more of a loop than my preferred (rotated layout), but it could still work.

I always thought you could just have a map end facing north and start facing east without any trouble, but I guess it’s not that simple.

So to clarify, despite what is in your image 2 and 3, this is what’s going to happen:
-you are NOT rotating C against B or B against C
-you are ONLY rotating the chiral center of B (e.g. the rocket and all platforms, scaffolds associated w/it) such that it aligns w/C had it actually been rotated 90 deg.

If this is the case, then this whole chirality issue is essentially a moot point. This issue ONLY emerges if you were LITERALLY going to rotate the B and C maps against each other, not create the ILLUSION of doing so. I brought it up so that you could cover every base regarding spatial orientation because this was the ONLY remaining issue preventing a perfect remake. Rotating the moon and rocket in B is the solution to the problem that emerges from the solution (90 deg rotation of B ) of the original problem (narrow travel arc between B and C w/o rotation), not the solution to the original problem itself, if that makes sense. If you’re not going for a LITERAL rotation of the B and C maps, which is the ONLY true fix to this that also makes Apprehension realistic, then you might as well not deal with the chirality issue at all since your map won’t be spatially accurate. Gameplay wise, the player probably won’t see shit anyway.

Edit
What targ is saying w.r.t. editing the end of B so that it’s not too close to C is exactly the same thing that I’m trying to achieve w/rotation of the two maps. However, the problem w/targ’s solution, mine and probably everyone else’s is that none of this shit matters unless you are ACTUALLY going for a spatially correct remake. It’s like a meta-problem.

I had this idea aswell, cause I guess u can warp from one place to another place at the end of each level like u said

U end North and u can start South through those warp during loading, cause it loads another map, and nobody will ever know u just start in a complete other direction

Wait. Why can’t you just add a turn to A1? That way, you don’t have to rotate B by modifying B, because if you map A, A1, and B together, B will be rotated already because it’ll connect to A1. The only people who will ever know that B is technically not rotated the right way will be people who look at the map files in Hammer or whatever, not people actually playing the game.

@Drag, Someguy
You can’t do the simple shit you’re talking about because you see the SAME ROCKET at 2 points…which means the path LOOPS back on each other. This is what’s creating the map overlap problem and the symmetry issues.

@Max
this has nothing to do w/A1. The problem emerges from BREAKING B and C to slot the new map.

No, I mean, can’t you practically rotate B without actually rotating B simply by adding a turn to A1 such that the player’s entrance into B is rotated?

I don’t know if you’ve ever taken organic chemistry, but this area is NOT the problem. It’s like in molecules where the chiral center determines handedness in the molecule. The area you are describing is like an achiral center away from the problem point which renders your solution irrelevant.

The PROBLEM is w/the rocket itself because you can see the frame of reference of C in B and vice versa. Rotational shit that you have to do MUST be here to correct the issue. Any other type of rotation ANYWHERE ELSE is irrelevant.

Hmm, I think I get what you’re saying. So the problem strictly revolves around B1 and B2.

I think the best solution is to do what Text wants (ie, just rotate the skybox and related features), but to also add a turn to A1, to avoid north spontaneously becoming east while the player isn’t looking.

Thinking about this is seriously doing my head in. It’s not my strong point. But, it needs to be thought about I guess.

I just went back and did a FAR MORE accurate composite map mixing B with a rotated C, and an accordingly placed Apprehension. I was very careful this time to align everything vertically and horizontally as accurately as possible (which admittedly is nothing like as accurate as I’d like - it’s difficult to manipulate ENTIRE MAPS accurately). It helps to conceptualize it on vertical levels.

B’s tunnels are on levels 1 (map start) and 3 (map end, after priming). C’s launch pad is on level 4, the drop down and tunnels to Apprehension are, very luckily, on level 2. Even though B and Apprehension overlap horizontally, they DO NOT overlap vertically, thus there is NO contradiction! Well, that’s not strictly true, there is a tiny, absolutely insignificant spatial contradiction.

Remember this room?

There’s a small grate on the floor, and beneath that is a flooded, inaccessible little gap for the pipes. This clips into a tunnel in B, by a very small amount. This is so insignificant and unnoticeable by ANYONE in game that I don’t care about it in the slightest. The overlap is highlighted. It’s ridiculously inconsequential, but I figure I’d best point it out, lest I be called a liar.

@Wangman - I think you’re perhaps not quite getting what I’m saying here.

I will rotate the skybox element and the rocket, now that I think about it, (the ONLY frame of reference for C’s orientation) in B by 180 degrees.
I will rotate the silo element (the ONLY frame of reference for B’s orientation) on C by 180 degrees.

This EFFECTIVELY rotates the C map 180 degrees, because the only 2 frames of reference (and they ARE the only 2) you have for its orientation have been rotated by 180 degrees.

Think of it this way - the map’s orientations in the map editor are actually irrelevant. You have to use common frames of reference to figure out how certain maps connect to one another spatially. For pretty much ANY Source map, that frame of reference will be the transitional area. You line up the transitional areas (with the rest of the map geometry selected) so that they overlap PERFECTLY, and you can then see if the maps themselves overlap, and how they’re relatively constructed. It doesn’t matter which way the maps were facing originally, it’s the RELATIVE orientation that’s the ONLY important issue here.

Because I’m planning on “disconnecting” B and C, the transitional area will no longer be a transitional area, and thus cannot be used as the frame of reference for spatial orientation. You have to use the frames of reference common to both maps, which in this case, are the ones stated above.

THUS, by rotating the 2 frames of reference, it EFFECTIVELY rotates the maps spatial orientation, again, effectively reorienting the C map simply based on those 2 common frames of reference (remember, the transitional area will no longer exist per se).

I’m pretty sure that makes 100% complete sense and I cannot find any fault with that logic. Maybe I’m missing something obvious and dumb. Let me know if I am. Using this logic I am currently CERTAIN that this solution is BY FAR AND AWAY the best one.

B and C are the only 2 relevant maps here. A, A1, and to an extent B1 have no effect on this equation. The only way in which B1 has an effect is it needs a good place to live, but we have to mess around with B and C’s relative orientation for that to work.

Christ, this has done my head in. I think I’ve resolved it now. It makes perfect sense in my mind. Back to mapping!

@Max
Strictly revolves around B and C (i.e rocket platform+skybox).

If TEXT does the skybox and rotations now, it’s basically solving a problem that isn’t there. Who cares about the sky orientation if the entire spatial layout is fucked? It’s like noticing a small dent in a totaled car w/o realizing the whole car is destroyed.

Basically TEXT only has 3 options here:

  1. do what I said and use a LITERAL 90 degree rotation of C against B and adjust the rocket and skybox accordingly. This fixes pretty much EVERY problem that has been brought up and would result in an incredibly elegant remake, even if it’s a pain to do.

  2. not do what I said+slot new map AFTER B in which case the realistic spatial layout of the new section against the BMS sections would be completely fucked, but probably not noticeable gameplay wise.

  3. not do what I said+slot the new map BEFORE B which essentially allows you to ignore the spatial layout but catastrophically rearranges the chronology of the chapter.

Edit:
Hold on, processing…see new post.

Founded in 2004, Leakfree.org became one of the first online communities dedicated to Valve’s Source engine development. It is more famously known for the formation of Black Mesa: Source under the 'Leakfree Modification Team' handle in September 2004.