OaR - Loop Mod


Someone commented on the Workshop that the OaRU version’s not working for them - I went to try it myself, and I’m getting the same issue.

The game is still using the “stock” (OaRU version, I guess?) version of the first map, and there are missing textures everywhere. I don’t suppose anyone knows if there’s some way to set addon priority to fix this?

EDIT: Wait, fixed it - resubscribing to OaRU made things work again. Never mind!

EDIT2: Or not… Finishing the first map kicks me back to the main menu because it can’t find a connection.


You know how to fix OAR Loop Mod textures ?


I’m not quite sure what’s going on. Everything works perfectly fine on my machine. Can anyone show me some pictures of what is going on (the missing textures, I mean)?

Just for clarification, did all of you unsubscribe from the stock version of Loop Mod? You should have only the OARU version of Loop Mod if you have both of them.


I’m not subscribed to the stock version of Loop Mod, no. I can’t recreate the missing textures issue now, but what I did to fix it was resubscribe to OaR Uncut itself - for some reason, it seems like the addon priority for Black Mesa is that older addons take priority over newer ones.

To reiterate:
Subscribing to Loop Mod while already subscribed to OaRU - game uses OaRU’s version of bm_c2a2a, and there are some missing textures at first (the one that comes to mind is the office window on the angled wall)
Subscribing to Loop Mod, then subscribing to OaRU - game uses Loop Mod’s version of bm_c2a2a, no missing textures appear, but the game returns an error that it “can’t find a connection” between c2a2a and c2a2a1. Using the map command to go directly to c2a2a1 and onward shows that the rest of the chapter is using the OaRU content.

I don’t know how or why, but I suspect the missing texture issue is fixed by subscribing to OaRU after Loop Mod - from there, it doesn’t seem to matter what “order” they’re in.


Welp, turns out vpk didn’t get the exact files I wanted it to. But I can confirm that the level transition is working now.


Kinda late, but yep, can confirm it’s working now! Thanks!


Bumping this in order to share an update.

As you might have been aware, the mods have not been compatible with the team version as of late. The recent update to OARU brought it to my attention but chances are the stock version was broken even before that.

Nevertheless, I think I’ve fixed it.

The sad truth is, ever since the transition to Source 2013, I haven’t been able to compile the maps at all. I never said anything about it even when porting the mod over to the workshop (I mean, it worked decently enough, after all), but wasn’t sure what I was going to do. And to explain why, the sdk tools have a built-in brushsides limit (which I previously dealt with during Loop Mod’s development) of 65536. In SDK 2007, Loop Mod was just under this at ~65300 brushsides. In 2013 however, as I recently sat down and did some tricks to find out, Loop Mod was about 72000 brushsides.

Why the jump? Well, compilers in different engines probably cut the maps in different ways.

Either way, I knew about the broken maps for a week and the reason it took me this long to get the fix out was contending with that brushsides problem. I used every trick that I know in tackling this problem; marking some areas for Propper and making some small detailing changes that simplified geometry to boot, etc. I was able to cut about 3500 brushsides. Sadly, that was only half of what I needed to cut. Anything more than that would have probably required some fairly radical changes to map design, which wouldn’t happen.

So I bit the bullet and compiled the map under 2007’s SDK where I knew it would still compile. And, now, it’s uploaded on the workshop. With any luck, you shouldn’t be able to tell that it’s compiled any differently (in fact, if you’ve played the steam versions already and not noticed anything odd, then you’ve already seen what I mean).

I did some rudimentary testing on these newly compiled maps but I’m pretty sure that they work. Let me know if something is still catastrophically broken.

Phew! Now that this is out of the way, I can work on updating Vent Mod tomorrow, which should take only tomorrow. For now, cheers!


(This post is slightly companion to a post made in the WGH - Vent Mod thread. Post here: WGH - Vent Mod)

Hello, everyone.

So I’m currently working on an update to Loop Mod. Specifically, I want to get the map into a state where I can compile it in the newest source engine. Currently versions of the mod use a map compiled under BM’s 2007 version. While I don’t readily know what kinds of issues (if any) this causes, I do figure that there is probably a drop in quality in the final map.

I am somewhat interested in having consistency, especially since I know, for a fact, that the 2013 version is much more rigorous in its compilation than the 2007 version and a Loop Mod compiled under it would probably be so much better.

However, actually realizing this will be difficult and has already forced me to compromise one of the main design principles that I had with the mod. I have time and time again shown pride in my ability to keep the stock areas of the map completely intact and I have stated my intention of keeping things that way.

Sadly, I have to face the truth that, in order to get this map working in the 2013 version, I don’t have a choice anymore. I’ve previously talked about how the map has gained several thousand brushsides on transitioning from 2007 to 2013, and this limit is not something I can simply circumvent. In the past, I have taken great strides to simplify geometry, use Propper for especially difficult polygons, and many many other tricks to get that number down, and it worked in 2007. But I am out of options.

As such, there will be some changes to stock areas. I will mainly be focusing on areas that are not frequented by players, things which should have little to no impact on the average playthrough. I’m trying to tread as lightly as I can, and I’m not sure how much more I’ll need to do before I can call it good. But bear with me.

I also wanted to let you all know that I am currently exploring options on making Loop Mod easier to work when other mods are present on the same machine (and vice versa for other mods). I can’t promise that anything will actually come of this but, since I’m making this post now, I figured that I would say something.


It might be helpful to make similar posts on the mods respective workshop pages.
Then again you might wait until they are updated.

EDIT: You already are :stuck_out_tongue:


Hey, everyone! Loop Mod has finally been updated into the newest engine.

I’m happy that I’ve been able to do this. The inability to compile in 2013 has haunted me for the longest time and I’ve just beaten it. To contextualize this, when I first tried porting Loop Mod into the source version of Black Mesa, I had 7000 brushsides more than what the compiler would allow. My task was to bring that number down. After using pretty much every trick I could think of, I was able to knock off about 3500 brushsides without having to actually change anything (as was my former design principle). I then hit a block and never thought I would be able to knock off another 3500 and thus get to a compilable state. The maps still worked in the 2007 compiler!

But I never thought I would be able to get it into 2013, and here I am!

It has, however, come at a cost. As I mentioned in my previous post, I stated that I was going to have to make some significant changes to existing areas. This included my own areas as well. But also as previously stated, I’ve avoided anything that should impact gameplay to any appreciable degree. These changes are as follows:

  • Simplification of the health charger room to the left of the first elevator
  • Removal of stairs in the ramp room
  • Removal of lowground near the electrical box room (where two vorts are fought)
  • Simplification of the H room between the two tracks where a bullsquid is fought
  • Simplification of a lot of things at the electrical pool puzzle.
  • Deletion of some vents and pipes around the map.
    At present, I’m sitting on the cusp of where the map stops working; the difference between the map compiling and not compiling is, in fact, a single brush. I’m that close. So I can’t stress enough that I really didn’t have a choice in these matters but to get it into the state it is now.

With any luck, however, the maps should play the exact same. Hopefully y’all will agree with me.

That said, I did want to talk about a major development that helped me loads.

You see, the way that Black Mesa handles tunnel curves is very expensive brush-wise. I would suspect that a large number of brushes and brushsides comes from how those were constructed. I would dare say it’s inefficient. They’re also quite modular in a way and, considering that the stock maps weren’t as full as this one, it was a non-problem

But what I attempted to do was rebuild the tunnel curves from scratch. To do this, I took a length of tunnel and rotated it around a center point (which would be the radius of curvature). I did this in 10 degree increments, giving me 8 portions of track within the curve (just like BM uses 8 portions of track in the curve) plus 2 more sections that lie on-grid. This works best when these two on-grid sections are connected right with surrounding sections of tunnel (especially if said sections are straight). The main difference, however, if that the wall cladding, the track edges, and the sloped parts of the ceilings, which are expensive in the old way, are done much more efficiently.

To illustrate this, a single piece of sloped ceiling is made of 2 brushes of 5 brushsides each, and there are two of these pieces; one on each side of the tunnel. The wall cladding is, similarly, made of 2 brushes of 5 brushsides each, and there are two of these pieces; one on each side of the tunnel. The track sides are made of three brushes with one being 6 brushsides and the other 2 being of 5 brushsides each, and there are two of these setups; one on each side of the tunnel. So, in total, that’s 2x4x2 + 2x5x2 + 2x5x2+2x1x6 = 72 brushsides. Multiply this by 8 curve segments and you get 576 brushsides per curve.

In contrast, with my method, the cladding is made of a single brush of 6 brushsides (and again there are two of these), the sloped ceiling is made of a single brush of 5 sides (are there are two of these), and the tracksides are made of a single brush made out of 6 sides (and, again, two of these). SO we get 1x6x2 + 1x5x2 + 1x4x2 = 30 brushsides. Multiply this b 8 curve segments and you get 240 brushsides per curve.

You can see how helpful this is in overcoming the brushsides issue. It turns out that I was able to employ the method on four curves in the map (and I briefly considered using it on a fifth). I think this actually saved the map, if I’m being honest. As an added bonus, however, the curves came out looking really nice.

So there we have it. The mod has been updated. Enjoy!


Have you considered using a tool like Propper to convert some of the expensive sections into static props?


Of course I did. As a matter of fact, during the alpha phase of the project, I had to use Propper not only on that electrical box in that side room there, but I also used it for the bridge at the electrical pool puzzle. Among other places.