Announcement

Collapse
No announcement yet.

So Epic....Lets talk about BSP

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • replied
    Nice rant you've got there. Simplified, well summarized, too bad no one gives a s**t...
    I appreciate the SuperGrid plugin, but a plugin like SuperGrid shouldn't cost any money. It shouldn't even be a plugin.
    What kind of person prefers an asset placing tool over an actual level editing tool?
    What the hell happened???

    Leave a comment:


  • replied
    Oh, I hope I won't look spammy >.>
    I find BSP pipeline a little bit clunky so I made a tool for rapid prototyping, it's quite popular in UE4 audience, so maybe some of you will find this useful too.
    Here is lil old video and link to UE4 thread


    Aand again - sorry for shameless plug :<

    Leave a comment:


  • replied
    I'd definitely also like to see nice robust *call-them-what-you-will* tools. I'm finding it incredibly clunky compared to UEd 2 and 3, and extremely frustrating. I'd very much second that with motion and flow being so important, having to swap between applications is a way to slow progress up immensely - ok, with experience you're going to know much better whether a 60m high nave "feels" right, or whether it doesn't, but to be messing around to and fro between a 3d application and the editor to get it feeling right is a headache that a lot of people don't need.

    I love UT, and while I never completed a level to release standard, I'd hoped to finally do that for this version. But the reality is that I'm a physically decrepit, middle aged parent with too many demands on my time to be able to learn a 3d application thoroughly. All I want is decent geometry tools and some nice meshes and materials. Accessibility is going to be extremely important if we want people actually mapping for UT - it's worth noting that about a year ago, I taught my then-7-year-old to do a CTF map for 2K4 in a couple of days. I'd really like to have that level of accessibility and usefulness in geometry tools for this version too, because I want to see plenty of maps (just maybe not by 7 year olds, they tend to have rather strange ideas about what textures go well together and what amazing juxtapositions you can make of castle meshes with space age ones )

    Leave a comment:


  • replied
    1200 views in 4 days on my first thread!?

    Holy cow. Hot topic indeed, lol at the very least I've piqued peoples interest. I've received quite a few warm and welcoming sentiments in both this thread and private messages. I never expected my first post on this board to get such a spirited response, thank you UT community.

    So first, I wanna clarify when I use the word BSP. Its this catchall idiotic shorthand that I've used in my head forever, and it seems its caused some confusion lol. When I talk about "BSP" solutions - I really mean just about any solution - Cross sectional geometry (CSG), binary space partitions (BSP), runtime booleans, mesh modeling solutions like Unitys probuilder, weird-*** voxel voodoo - just about anything that promotes this type of in-editor workflow. I know, I know, BSP as a catchall term makes no sense lol, its a very specific thing. The failure of communication is entirely on my end - I used lazy terminology like a lazy son of a gun, and it appears to have confused the issue a bit. My apologies . Err what word should I use henceforth? "In- Editor Prototyping Tools"? "Geometry Tools/Solutions"? Whatever term is the best catchall, Im open to suggestions.


    Responding to a couple of the replies since I've last posted:
    Originally posted by [Epic]Chris.Kay View Post
    I'd like to chip in here with a few words. Our BSP/Geometry tools need work for sure, we all agree they can be improved, I'm not sure when or if it is planned though. As others have mentioned, it is useful for creating shells for a game such as UT, the bugs in the current BSP system should be addressed in due time like any other bug from what I understand (material alignment issues etc)


    I've used BSP/Primitive editing solutions from all the major engines that have support for it, and I managed to adapt quickly to Unreal, suppose I've got used to various limitations in each toolset, the "shell" after all is perhaps only a quarter of the work that goes into a finished level.


    Yes, 3d packages are a great solution that I enjoy using, but I would only use it if the level called for it, vast open terrain, organic structures (like a tree house theme, volcano etc)


    Obviously I'm referring to UT specifically, there is a such a variation in workflows for other games and companies, we should keep this focused toward UT.

    Yesssss! Thank you for participating Mr. Kay. Im ecstatic to hear a UT dev weigh in. Id agree regarding keeping the convo geared toward improving the workflow for UT, its a fantastic idea - and may help focus the conversation a little bit.


    So you say you agree BSP/Geometry tools need work. You also mention bugs in the current system that should be addressed in due course - do you imagine Geo 2.0 to simply be a faster, bugfree version of the toolset we have now - or more of an overhaul like Niagra?
    What similar systems have you used that you've felt had some great additional functionality?
    Disregarding feasibility for a second, What is/are your dream solution(s?) - if given free license to speculate on exactly what you would want out of our toolset?

    Like I've said in previous posts, my goal here is to kinda crowdsource brainstorm what kind of improvements are desired/realistic - and maybe generate some community momentum to help get Geo 2.0 promoted out of the wishlist/backlog.


    Originally posted by MoxNix View Post
    Going slightly off topic but still closely related, the 3 main things I'd like to see change ASAP with BSP editing in order of importance are


    1. Fix the texture alignment bugs.
    2. Make the Lathe tool work.
    3. Give us the ability to convert static meshes into brushes. I found that real handy in the editor for 2k4 when I wanted to make simple changes to a mesh like vertex editing it to be wider or taller without deforming other parts like scaling along a single axis can do, changing, realigning or adding more textures. All I did was convert to brush, make the changes, build and convert back to mesh. Quick, real easy to do and it worked great!

    Its not off topic at all. Its literally exactly what I asked for: how would YOU like to see this tool improve? If anything I'd like to see more people pop off with improvements they want and what they'd like Geo 2.0 to be. How do we improve our UT level design toolset?


    I feel like im hearing alot of "BSP sucks, just use a 3d modeler" - which is fine, 3d application-centric workflows will always exist, and its a sentiment that has been clearly established in this thread....however it'd be nice to transition the focus of the convo to constructive feedback on the tool (it seems like peoples posts are naturally gravitating towards this anyways, now the topic has grown a little). Someone giving their thoughts on the issue, and a list of improvements they'd like to see, is far more constructive all around. And arguing amongst ourselves over what improvements are viable and realistic is more constructive then the BSP vs 3d application debate, which is a debate that will go in a circle forever.


    Im sorry to the posts I havent replied to, if that matters to some of you. I read em all. The conversation, in general, seems to be making nice progress with or without my input. Theres way more response then I was expecting, replying to all is....overwhelming, and I feel like weighing in on each and every response is a bit presumptuous on my part. Like im some Czar of this convo or something. In time, I'll scour through the community and dev responses and update the original post with the ideas and constructive feedback you've all shared thus far.

    Keep up the feedback!

    Leave a comment:


  • replied
    BSP/CSG are related somehow? CSG is the.. object and BSP is the method of storing/looking up stuff for the geometry? Or something. It has always been this way?

    Leave a comment:


  • replied
    Yeah, so people should stop referring to it as BSP then...

    https://en.wikipedia.org/wiki/Binary_space_partitioning

    Leave a comment:


  • replied
    It hasn't been BSP for a long time. I think it's actually called CSG (constructive solid geometry). Not really sure what the difference is.

    Leave a comment:


  • replied
    Originally posted by Tidal Blast
    Look, I was simply trying to point out that level design as so much to do with art that it just shouldn't be ignored. Level design just cannot be a specialization on its own when it is actually the combination of both environment art and game design; a multi-specialization. The best level designers that I've met or seen in the industry were once environment artists who later became level designers, which is obviously a very natural path.
    I'm not saying I disagree, but I think that kind of approach applies way more to single player level design. There, you'd use art and lighting to direct a player through a level. For a game like UT this isn't really the case, since the game is fundamentally about a player's interaction with a layout. Using light/art as part of the "gameplay" is much more of a special case than the norm for most levels. Being both an artist and a level designer at the same time will probably cost you time, because you'd have to redo art to suit changes in the layout with that sort of workflow. Level design for a game like UT is much more about the intricacies of a layout that will be played again and again and to a certain extent needs to be memorized by the player. For those kinds of projects you need to be much "closer" to the game in a stripped down state (ie, layout shells with no art) in order to make small iterations to a layout. Jumping back and forth between two or more tools and constantly exporting/importing is a serious time sink. This is why I think your instance on workflows outside of the editor is misguided. If the in-editor geometry tools were solid, there would be no benefit in using Max or whatever to create layout shells.

    Originally posted by [Epic]Chris.Kay View Post
    I'd like to chip in here with a few words. Our BSP/Geometry tools need work for sure, we all agree they can be improved, I'm not sure when or if it is planned though. As others have mentioned, it is useful for creating shells for a game such as UT, the bugs in the current BSP system should be addressed in due time like any other bug from what I understand (material alignment issues etc)

    I've used BSP/Primitive editing solutions from all the major engines that have support for it, and I managed to adapt quickly to Unreal, suppose I've got used to various limitations in each toolset, the "shell" after all is perhaps only a quarter of the work that goes into a finished level.

    Yes, 3d packages are a great solution that I enjoy using, but I would only use it if the level called for it, vast open terrain, organic structures (like a tree house theme, volcano etc)

    Obviously I'm referring to UT specifically, there is a such a variation in workflows for other games and companies, we should keep this focused toward UT.
    Do you know if the "BSP" in unreal is even actual BSP at this point?
    Last edited by Quotidian; 07-07-2015, 02:42 PM.

    Leave a comment:


  • replied
    I personally love BSP. I am a mapper that has been previously focussed on UT2004. During that time I have made maps that have too many BSP brushes for UE4 to even handle (it crashes when I try to import it). I released maps under the name ChookWoods in UT2004. These maps were near pure BSP and they were also pushing the limits on the number of brushes used. Whilst I appreciate I cant get away with making a pure BSP map in ut4, I also understand the need for it to exist, and to exist in a sense that is highly usable.

    Leave a comment:


  • replied
    Nah they are just hidden in menu tree things. Gotta expand them.

    Leave a comment:


  • replied
    Speaking of bugs, is the postprocess volume screwed this release? I can't see any options beyond the basic brush settings.

    Leave a comment:


  • replied
    Going slightly off topic but still closely related, the 3 main things I'd like to see change ASAP with BSP editing in order of importance are

    1. Fix the texture alignment bugs.
    2. Make the Lathe tool work.
    3. Give us the ability to convert static meshes into brushes. I found that real handy in the editor for 2k4 when I wanted to make simple changes to a mesh like vertex editing it to be wider or taller without deforming other parts like scaling along a single axis can do, changing, realigning or adding more textures. All I did was convert to brush, make the changes, build and convert back to mesh. Quick, real easy to do and it worked great!
    Last edited by MoxNix; 07-07-2015, 01:51 AM.

    Leave a comment:


  • replied
    I'd like to chip in here with a few words. Our BSP/Geometry tools need work for sure, we all agree they can be improved, I'm not sure when or if it is planned though. As others have mentioned, it is useful for creating shells for a game such as UT, the bugs in the current BSP system should be addressed in due time like any other bug from what I understand (material alignment issues etc)

    I've used BSP/Primitive editing solutions from all the major engines that have support for it, and I managed to adapt quickly to Unreal, suppose I've got used to various limitations in each toolset, the "shell" after all is perhaps only a quarter of the work that goes into a finished level.

    Yes, 3d packages are a great solution that I enjoy using, but I would only use it if the level called for it, vast open terrain, organic structures (like a tree house theme, volcano etc)

    Obviously I'm referring to UT specifically, there is a such a variation in workflows for other games and companies, we should keep this focused toward UT.
    Last edited by Chris.Kay; 07-07-2015, 12:14 AM.

    Leave a comment:


  • replied
    Originally posted by Tidal Blast
    Look, I was simply trying to point out that level design as so much to do with art that it just shouldn't be ignored.
    Rightio. Was hoping you would be able to shed a little more light from what I have assumed you meant but I guess "learn 3D" is enough.

    I think this was the most relevant image from the post I quoted.
    Last edited by joellll; 07-06-2015, 09:44 PM.

    Leave a comment:


  • replied
    Originally posted by Tidal Blast
    Many years ago, game level creators were both designers & artists. At some point, game studios started to split the job into multiple jobs and now we have level designers and environment artists. Essentially these days... level designers create a blockout, test gameplay for two weeks, then artists arrive, they break down the level to create modular assets, they'll change some areas of the map to fit the art, they'll create unique assets to fit the blockout, etc. However, when it comes to the creation of game levels, the art is the bottleneck and there are many issues with the method that I just described. Designers waste a lot of time testing the gameplay when some areas will change anyway when artist will touch the map. Artists waste tremendous amount of time trying to break down the map. Artist will have to create more meshes than they should to fit the custom blockout, etc. The workflow that Hourences uses is a lot more efficient. He creates the modular meshes first and then build the level with those. Right from the start, that means that he doesn't create a blockout and waste his time trying to adjust the gameplay when it will change later anyway. Because he uses meshes to build the level, he doesn't have to break down a blockout into modular meshes, etc. He basically works at least 2-3 times faster than usual, plain and simple. And because most level designers these days aren't also 3D artist, the quality of their map layouts often take a hit.
    This plus the post below and your references to hourences indicates you want pieces of the level that can be snapped together to create the layout? Are you suggesting people create these modular sections before starting their layout? Or are you suggesting that level designers work on a base grid (50, 100, whatever, in the 3d package) so environment artists can mesh the level more efficiently using blocks like this?

    https://forums.unrealtournament.com/...l=1#post148402
    https://www.youtube.com/watch?v=8SxTqlom3Fg 22minutes

    Why is much time wasted breaking down the map to mesh it? Can you explain how using a 3D package solves the custom blockout problem?
    Last edited by joellll; 07-06-2015, 08:36 PM.

    Leave a comment:

Working...
X