Blender 2.91's new Boolean system is INSANE

hey guys i want to discuss blender's crazy new 2.91 boolean system now we're still in beta unfortunately and i tend to not use betas but a lot of people are bringing this up to me as well as with the new release of hard ops i figured i would discuss the new system so this thing is crazy let me go ahead and show you so traditionally in blender whenever you try to run a boolean on coplanar geometry meaning the geometry occupies the same point in 3d space let me show you kind of like this for example these two top faces are at the same point on the z-axis there were issues when it came to using booleans so there's two types of algorithms that have been implemented in 2.91 fast which is the one that blender has been using forever and exact which is the new boolean system now let me show you the original issue with booleans in blender so in 2.90 before this new boolean system was implemented if you try to run a union boolean on coplanar geometry like this you're going to see it simply doesn't work so usually to fix this what you'd have to do is make sure the boolean isn't running on coplanar geometry and usually you can just come in here and slightly scale it up on the z-axis just a bit and that's going to actually fix the problem as you can see here now check this out if we go into 2.91 and use the new boolean system you're going to see that this actually works right out of the box on coplanar geometry no need to scale up or down it works right out of the box as you can see and there are absolutely no problems and this is because of blender 2.91 new exact solver which we have two different ones right now we have fast and we have exact this is just in the hard ops helper menu for quick access so if i were to change this to fast instead of exact you're gonna see the issues are actually gonna come back right here so now if we like run the bevel same issues as before it kind of goes crazy but if i swap this back to exact it actually works pretty fine no real issues i can just run a hard normals and we're pretty much good to go so there is one downside i don't know if it'll be you know fixed out the box later on once again we're still in beta not really a big deal but if we go into wireframe view here you're gonna see that we still have this random edge kind of sitting on the top we don't want this so usually what you need to do at least for now we need to hop in here and add a decimate modifier set to planar we're going to put this above the bevel and that's basically going to clean up that extra geo sitting on the top so i'm hoping that maybe in future releases this will be able to just kind of be done automatically so i don't have to get in here and clean that up it's not really a big deal but it does require an extra step most of which we want to avoid modeling is all about efficiency so one other issue i've seen is with the difference boolean and this one isn't really an issue at all for me and i'll show you why so if we were to go in here and try to do the same thing but instead we use a difference boolean instead of a union maybe i'll come in here right if we come in and try to run a difference boolean sometimes it works sometimes it doesn't in this case it was actually okay but i found another situation sometimes it doesn't really work as expected like you can just see there is a glitch there right now it seems to be working all right for me but you know in some instances it kind of pops back into place and you have to scale it up on the z-axis a bit so that's one issue i've ran into so i'm hoping that'll be fixed as well but as you can see this is a courtesy of the fast solver if i get this in a good position let me find a good one that's actually working see now we're actually running into more problems and it's probably worth here we go so if i go over here you're going to see we're still set to exact but if i swap this one over to fast basically it's not going to work anymore so there's obviously still some clear issues with this new boolean system sometimes it just doesn't work as you would expect and obviously we're still in beta so you can't really be surprised about it but this is insane because we don't have to bother going in here and scaling up the axis anymore and you know dealing with that whole spiel because in difference booleans it's okay you can just scale the axis up as high as you want and it's going to have the same effect but if we were to run union booleans that's an issue because if i start scaling this up then we have kind of uneven geometry as you can see in this example so really i find this more practical for union booleans because i want to make sure everything is completely coplanar for difference booleans i don't really care too much but it's nice to have the system where it runs a difference just fine with coplanar geometry now i have been using the hard ops helper menu for this but the same exact settings are available to us here in the boolean menu i just don't like having to go in here every single time and change it from fast to exact or whatever so that's about it for the new boolean system it is amazing i don't even know how to articulate how powerful this thing is going to be once it's fully functional i just wanted to give a quick preview of the beta version and what we have available to us now so thanks a bunch guys for watching hopefully this gave you a bit more insight to the boolean system and that's about it thanks for watching and i'll see you in the next video

Channel: Josh Gambrell

Views: 129,918

Keywords: blender, 2.91, boolean, algorithm, update, system, chipp, walters, hard, surface, bool, booleans, tutorial, difference, union, coplanar, geometry, topology, merging, hardops, boxcutter, josh, gambrell, non, destructive, modeling, 3d, 2.9, 2.8

Id: GDILtiN36AA

Length: 5min 12sec (312 seconds)

Published: Tue Oct 27 2020