Welcome to Keen Software House Forums! Log in or Sign up to interact with the KSH community.
  1. You are currently browsing our forum as a guest. Create your own forum account to access all forum functionality.

Round Ledge Battlements don't snap into walls, a z-fighting idea.

Discussion in 'General' started by Tenzo, Apr 18, 2019.

Thread Status:
This last post in this thread was made more than 31 days old.
  1. Tenzo Apprentice Engineer

    Messages:
    259
    Hi,

    So I just noticed that the ledge variants of the battlements, don't snap into walls that are placed just on the exterior of another piece. For example, walls placed back to back, form a thick wall. Well, the round ledge battlements for some reason don't want to snap into place where there is a double wall. They just don't like thick walls, which means you can't connect them unless you strip a layer of stone off your castle. If you have connected double layers just to reinforce structural integrity, that might not be possible for your design and you won't be able to connect the part of the design that is using the round ledge battlements. You'll have to strip them down and use something else to connect them.

    The round battlements connect to thick walls just fine, so I was thinking, maybe the ledge battlements should also work the same way.

    Round ledge battlements are vanilla pieces, so you want them to have the most functionality within the existing block set because it's something you can control.

    It doesn't make sense to mod ledge battlements in order to snap into exterior walls. All players should have that. It's a vanilla piece. You want them to have the most versatility before the player finds themselves having to resort to mods in order to finish a design.

    The Machicolations also have this problem. They are a one trick pony. I hope you guys think about extending their functionality because they are vanilla pieces.

    Ledge Battlements also don't connect with singular arches. So if you want to make an outer bridge between two castle pieces, you have to do regular battlements, then when you can do pillars, only then you can connect them to ledge battlements. But I think if you wanted to, you should be able to make a straight bridge with ledge battlements. Which means they should also be able to snap to singular arches, like their smaller cousins.

    In order to solve z-fighting, I really think you guys should have a system of detecting edge collisions and move the straight edge of the piece in the "placement mode", back by 0.001 units. That way, you will eliminate all cases of z-fighting while being able to be more fluid with block placement.

    Also, wooden stairs don't connect to a wall built on the exterior, even though they have plenty of space to accommodate for the wall. Regularly, you have to put a plank in front of them just so you don't step on empty space when you connect them to floors and a wall on the outside. So, why don't they snap in a more friendly manner? I hope you guys check it out, there is so much space there. There is no reason why they shouldn't be able to compound with an interior wall setup.

    Also, large round floors, do not connect in any way with large round battlements. You can put a large round wall on top, but not their battlement variation. Even though they have the exact same geometry underneath.

    There are a lot of inconsistencies like that throughout the existing set pieces. I hope you guys can find a smart way to solve their functionality issues. When you place blocks, you can enter a temporary edit mode where you can try to adjust their edges in order to not have exactly identical texture locations, so the textures don't fight for which is in front. You can address texture z-fighting without making it the player's problem by using smart solutions, not 50 different compound rules.

    Which in the end, will be a much better solution and eliminate your mount point headaches, to which there has been no end to, ever since the "intuitive block placement" solution, which was not really a solution - it just introduced a new set of problems. And it put them on the end user to solve. You guys just redid the grid system.

    With a fresh memory of the code, please think about having the system think about matching planes and just slide one of the planes back by 0.001. Simple fix, and it won't make it egregious on the players to solve placement issues because of z-fighting. Please think about that.

    Even after the intuitive block placement solution, z-fighting is still alive and well in the game, with certain block combinations. It has not even been eliminated by this current setup. But think of the hassle both player and developer have gone through ever since its introduction. No end in sight to the problems of valid and invalid block placement. I won't report the blocks in question because I am afraid that will just end up in them being nerfed by not being able to connect anymore. (Even though they have perfectly valid connections. It's not a cube on top of a round corner for example.) I don't need more of that when I'm building.

    Yet we have to deal with the invalid block placement issue, even though z-fighting still exists.

    You need a different solution. Or forever you will remember your ME development days as solving "mount point issues". You won't remember that you developed a great game, you will just remember that hassle. And so will the player. Yes, I just built a great castle yesterday, but don't ask me the pain I went through because of the same "mount point issues". So, if both player and developer have the same problem, and z-fighting still exists, we need a different solution. A smart solution that works for both parties and doesn't lead to you forever iterating on each block combination, each time you add a stone block, from now until the end of time.

    Or mods for vanilla blocks that should connect, but don't connect.

    And the endless stream of reports that stem from that. We'll never get to actually develop the game.

    Please, think about that. Both end user and developer can benefit from a smarter solution.

    How many more years of "hey, these blocks should connect but they don't", "yes, you're right, we'll fix it" reports can you take before you crack?

    Enough is enough. Seriously, mental health is an important consideration when you go into work each day. Protect the sanity of ME developers by recommending them smart solutions when you post your own block placement bugs.

    Think of the years we have been though with this issue alone. It is the biggest source of complaints and bug reports in the entire ME development history. If it is not addressed in a smart way, it will continue to remain so from now on until the dawn of time.
     
Thread Status:
This last post in this thread was made more than 31 days old.