Agreed all round.
Also, HammUEr is out! The best stopgap you’ll get until Epic make 2.0, plus you can port all your old Quake/HL/Source levels.
[=;442774]
Unreal’s current BSP system really needs to be deprecated entirely and replaced with a new system, which I believe is the plan. Which is great, but it’s probably going to be a while 
[/]
Heck, no.
Golden rule in programming is that “newer and shinier doesn’t mean better”.
BSP approach worked for years in other engines, and deprecating it completely would be downright silly.
“Completely new system” will take years and projects are being built NOW.
It will be better to just fix the bsps.
Or FIRST build a replacement and only THEN deprecate stuff.
Otherwise it’ll be like that situation with cascade. UE Answerhub says that cascade development is abandoned, except the “niagara” replacement for the cascade isn’t there yet.
Hey guys,
Long time lurker, first time poster.
Just to give a little context to my post, I’ve been working professionally as a level designer for about 7 years now across a few big companies. I’ve been working in my spare time with UE4 for about a year, just about to make the jump over to the indie side of the pond.
I’d love for there to be something faster than BSP for feeling out a space, but no such thing exists. I certainly can (and have) used things like 3DS to block out a space, but it’s akin to using a sledgehammer to crack a walnut. is even if we forget the fact that I’ve never encountered anything that can go from creation to playable faster than BSP.
I think there’s been a bit of a misunderstanding from some parties about exactly where level design begins and ends. When you’re designing levels all you care about is making the space work for gameplay. You may give some consideration for which areas will create vistas but you are intentionally not considering the art very much. You don’t care at all about the efficiency of anything you’ve created, how reusable is, how sensible any topology might be. I get the feeling some people think is lazy and ask ‘why not do it properly the first time?’. The most important thing a designer can do is iterate as fast possible, and spending any time making something neat when you might bin the whole thing is a complete waste of time and very disruptive to your thought process. Anyway, I realise that I am largely repeating what’s contained in the original video, but suffice it to say I agree.
For level designers, nothing is quite as good as BSP. The fact that someone has created a tool to port from other toolsets should be evidence that it’s something of value. Unreal has always positioned itself as a designer led engine and seeing people dismiss something like without the full understanding of what we are asking for is pretty discouraging. How important it is vs the other things in the engine is up for debate, but I don’t think there should be any debate that good bsp tools are still very useful.
I agree with . UE4 should have good geometry tools. Some years ago I had made some maps for Portal 2 in the original Source Engine Hammer Editor and it was super easy, intuitive and fast. It is very hard to do the same thing in Unreal Engine. If games like Minecraft can get popular, I don’t see why we can’t have good simple and optimized geometry tools.
Thanks for chiming in, Atomic. Really well put. I feel like a lot of people who think they know what level design is are mistaken 
As a level designer that’s competent at producing both BSP levels and creating my own modular meshes in a 3D package, I just cannot understand why BSP is so vehemently argued against by a small elitist minority who I can only assume aren’t level designers.
I can only see a lot of positives for having strong BSP tools around and currently UE4 doesn’t have them. I do hope they implement some as promised earlier in the thread.
Personally I love to prototype my environments in BSP first simply because of how fast it is to work with. I can really test important things that will alter the final aesthetics of the level. For instance the layout of decor in a room, the gap between pillars, how many pillars and how wide they are in that room. It may not seem important, but it can make a massive difference. For me, gameplay always comes before aesthetics in level design so I’m far more concerned with that than anything.
When you are a designing levels for a third person game with characters that have a wide variety of scales and abilities, you’ll very quickly appreciate the ability to be able to quickly change things to accommodate . Ceilings need to be at least high, doorways need to be at least wide and tall, there needs to be enough floor space to manoeuvre for combat etc. It’s all more important than how nice that fountain mesh looks.
The biggest positive of all for me however is how quickly the BSP workflow can be learnt/taught. That basis alone is more than enough in my opinion for it to be a better implemented .
Just some examples (by Simon “Sock” O’Callaghan) of what has been done with BSP: http://www.simonoc.com/pages/design.htm (except Crysis maps of course)
Here is another one:
Note that is was for Quake and Quake 3. Imagine how it can look in UE4 !!!
[=;443034]
Just some examples (by Simon “Sock” O’Callaghan) of what has been done with BSP: http://www.simonoc.com/pages/design.htm (except Crysis maps of course)
Here is another one:
Note that is was for Quake and Quake 3. Imagine how it can look in UE4 !!!
[/]
There’s no doubting that good maps can be made using CSG/BSP tools thats why UE4 needs its BSP tools improved and maybe the engine to handle the improvements, even if the level was build in BSP then converted to Meshes that would be of great help to the new comer to UE4 who’s use to level design in Hammer or one of the other tools and only using a 3d program for props or characters, Talking of converters i know HammUEr is out now But if you wanted to just convert >vmf’s to fbx models there’s Source FBX Tool and BSP Proper Tool - Old Developments - PixelTail Games - Creators of Tower Unite! i haven’t really played around with so don’t know how collisions are effected if they are
BSP has some advantages.
-
People not familiar with modelers or without skills will find BSP tools easy to use and they will be able to start making complex level design.
For example many Minecraft users are able to make incredible levels without knowing anything about a modeler tool. BSP is somewhat like Minecraft, very friendly to use and start making levels. -
They can workd directly in the editor without needing models import each time a modification is needed.
For example you have a game level with some street walls and you need to make walls with different height and some different holes, you just copy a wall and start editing it without needing to start a modeler and multiple import process.
If you look at Minecraft like games and all great levels total beginners are able to make, BSP is as easy to use and allows lot more than minecraft.
My god that hits the nail on the head. That was the first thing i had to wrap my head around was how awkward it is to manipulate the brushes in ue4. even after a year of using ue4 i still hate these barriers. im not looking to create a AAA title. Bsp’s are fine.
Ue4 is a fantastic tool but makes life difficult needlessly.
If they would implement these simple changes it would make the ultimate indie development engine.
The amount of bigotry towards BSP’s in thread is incredible, yet it’s one of the most important tools.
I have a feeling people don’t realize how important it is to walk through the level you design on the fly.
Voted, disappointed how little support there is.
EDIT:
You got some amazing stuff on your blog !
thanks! 
Hi, I found topic today and I felt connected to the majority in here.
Currently what UE4 provides, I don’t understand it and it doesn’t seem to be working for me. Whatever I found back in tutorials, the BSP just doesn’t seem to do that (I can’t even resize meshes with it other then using the Scale options in the Mesh properties).
It would be nice, actually bloody , to have a proper LevelDesign tool in UE4. But on the other hand, as we normally work in a 3rd party program, i.e. Blender/3D , it’s maybe better to continue with workflow. But back to the first hand, for quick prototyping for a level it’s a bit of a large workflow (model it, export it, import it, place it, find problems and restart the flow).
[=;464566]
Just popped my head in to see if had been fixed, and it seems like the card has been archived for some purpose on Trello. So I guess the new editor was trashed.
[/]
It’s been renamed for quite some time. It’s called Geometry tools 2.0
It’s bizarre to see a proposed which not long ago was practically an industry standard be reacted to with such hostility. Anybody who believes nobody wants good pseudo-BSP tools need only look at the popularity of ProBuilder in Unity. It’s one of the most consistently popular and highly rated plugins in the whole store.
Anyway, just chiming in to show my support. I’ll be purchasing HammUEr very soon, and I eagerly await ProBuilder’s upcoming plugin, but I still feel that having some kind of native, Epic-supported solution would be of huge benefit to many people either starting out or graduating over from older engines. Imagine the possibilities of something like combined with parralax occlusion mapping for rapid iteration. It could be amazing!
Unfortunately better brush editing tools would solve only half the problem, there is also a great performance cost if you don’t convert them to static meshes, which kinda undermines the whole point of having BSP’s in your level.
I’m not sure if there is a solution for it, but my guess is that’s exactly why it’s been in state for so long, because it’s too troublesome to optimize it, doesn’t mean it’s not possible though.
[=;464756]
I think, it’s because Epic/Unreal has a different methodology. People who have been taught/learned the “Unreal way” generally like that way.
[/]
That’s kinda funny to think about. UE1 / UnrealEd had one of the most, in my mind, elegant BSP solutions of any engine of its time. I’ve been jonesing for something half as intuitive ever since those days, and was really disappointed when Epic moved towards static meshes as a paradigm in UE2 onwards. That said, I understand it from an artist-centric perspective, but for a hobbyist like myself, it was never quite the same.
[=SimplyMark;464777]
Unfortunately better brush editing tools would solve only half the problem, there is also a great performance cost if you don’t convert them to static meshes, which kinda undermines the whole point of having BSP’s in your level.
[/]
I doubt whatever Epic comes up with will be a ‘true’ BSP system though, as yeah, it’s a nightmare for optimisation. My hope is that it’s some sort of system not unlike what ProBuilder does, where brushwork is automatically converted to meshes (at least I’m 99% sure that’s how it works), so you get the flexibility of CSG brushing, but none of the culling headaches. HammUEr’s system of splitting parts of the BSP into separate meshes upon import is also pretty neat. There are definitely workarounds. I don’t think anybody is asking for a full hardcore BSP implementation like Uncle Carmack used to make, but just something that makes rapid iteration a bit more feasible.
I really just want to bash bsp around like and then take it into max/maya once i’ve made something i like:
https://www…com/watch?v=2rS075Bp-E0
We could mock a level up with a styllus with kind of interactivity.
If you take a look at CryEngines Designer tools, its exactly what UE4 should have, even now to compete with the engine. Its basically a simple modeler inside the editor.
https://.com/watch?v=zqvhSPhY6EY
[=;473169]
If you take a look at CryEngines Designer tools, its exactly what UE4 should have, even now to compete with the engine. Its basically a simple modeler inside the editor.
[/]
Yes and no, because the 2D support is limited I think, the last versions Cryengine got tools, but too bad for 2D viewports, bad design.