4.17 Preview 1 ETA

Does anyone know when the first preview for 4.17 will launch, I am excited to try out the new geometry tools

Considering that 4.16 released last week, probably a couple months

You could always download the MASTER or PROMOTED branch on GitHub…


Will the geometry tools even be in 4.17? Not clear to me yet, since they arent listed for 4.17 on the roadmap at this moment in time. They are listed as ‘Polygon Modelling with Subdivision Surfaces’ in the Future Releases section of the roadmap, and I would expect that these tools also provide a good chunk of the foundation for another Future Releases feature, ‘Advanced Programmatic Mesh Creation’.

Since the roadmap description for ‘Polygon Modelling with Subdivision Surfaces’ speaks of an initial release with basic functionality they could make it to 4.17, I’m just saying that I would not make this assumption yet.

I would highly recommend anyone really interested in the geometry tools try it for themselves right now, no need to wait, just get the dev-tools-geometry branch off github.

I’m not sure they are in that branch, as per my previous post there is a completely separate branch for geometry tools and I dont think I noticed them being merged into master yet.

How do we know there is a completely separate branch? Roadmap? Like you said in your previous post, it just says that its a feature for future releases. Not a separate branch… Or, did I miss something?

Not sure what it would hurt (except for time) to download Master and see what’s in it…


By watching GDC videos or whatever Epic twitch stream it was in the last few months that Mike Fricker appeared on to go over the geometry tools stuff in more detail. Or by looking on github:


The roadmap will eventually tell us when its going to be available in a main released version but in the meantime that github branch is the best bet. It’s also possible to keep an eye on the commits to the main branch to see what is added to it over time, hence my comment about not noticing it going into main branch yet.

Or by reading the relevant docs:


I made a mistake in that I suggested MASTER / PROMOTED… There is a separate branch just for the Geo Tool. Source code is here: https://github.com/EpicGames/UnrealEngine/tree/dev-geometry

Apologies for the misinformation.


Ive tried that and loved it which is why I was wondering when the stable 4.17 release is coming :slight_smile:

You loved it? That, does surprise me - I found it to be clunky and very poorly designed. I think Joe Wintergreen said it best in his video here:

It’s not a modelling tool and not a level design tool.

Epic Games need to hire a level designer to design a level design tool, because right now, the mesh manipulation tool is an atrocious level design tool and an awful modelling tool. I don’t know what it is, truly. Perhaps a half-way house between the BSP tool and after converting it into Static Mesh, an opportunity to add additional detailing? I think that is where it sits right now.

It has created an additional step prior to passing the level over to the environment dudes to finish the level off.

if only the probuilder people could get off their butts and stop being lazy, and finish bringing probuilder to UE4 we would not have this problem

Yes but also keep in mind what Joe Wintergreen says in the description for that video:

Mike Fricker did say the non-VR interface was just a placeholder in a video quite a long time ago that I will probably struggle to find.

I’m not sure exactly what Epic are planning but I think one of the purposes of writing this stuff is to enable functionality under the hood that will be used for the other relevant feature on the roadmap, ‘Advanced Programmatic Mesh Creation’. I would kind of hope that 3rd parties will use the programatic stuff to build higher-level tools that are more like what people are after when it comes to level design features, perhaps that is what Epic hope too, I dont know.

Excuse the terse response, but I hope it doesn’t.

Epic should provide the ultimate level design tool, not dilute their engine on a whim.

I dont understand the engine dilution comment.

Anyway the reason I hope for that is that nothing resembling an ‘ultimate level design tool’ is on the roadmap. Things that can be a significant and solid part of the foundation for such functionality are very much cooking, you have to walk before you can run. And since Epic show no signs of being at the point of running with the sort of level design tool you are probably hoping for, and this thread is dealing with the more immediate future, I am bound to contemplate the significant chances of a period where 3rd parties build stuff with these pieces and add to them rather than hoping Epic will come up with a significant level design tool out of the blue.

And even in the short term, I still have no expectation of the geometry editing tools being in 4.17, I will be happy if they are but have no particular reason to get my hopes up. Looking at github commits for that branch shows refactoring & other stuff being done to it in the last 8 days. I havent tried these recent changes yet.

I may be reading wrong but i feel like some of you guys are still asking to turn UE into a full fledged modeling software to design your levels in? I think I have had this discussion once before in these threads.
I kind of understood it from a blocking perspective but in practicality never really made much sense to me over all to be usd in production. Why model in UE when you can do a better job outside the engine with better faster tools and just import?

If you are looking for a quick blocking to plan a level, I don’t understand why it wont be possible again blocking it outside the engine and bringing the elements in? Like many studios do even the very large ones.

Why so much on insisting on this feature when Epic can dedicate its time on the priorities, After all UE is not or never will be a proper modeling tool so why settle for a half baked one?

I just feel devs these days want everything faster and faster, more and more, all and all in one global tool. which is not possible and not fair especially when UE still lacks in many important areas that need improvement. The general workflow should be retained but simply improved. How about asking for connectivity plugins between 3d apps like Blender Max and Maya for faster and easier imports and exports of assets rather than in built unfinished buggy tools that don’t seem to belong.

ETA for future Engine releases can be found here: https://.com/c/UA40A4XN

Ah so that’s where i had the last discussion about this :).

I guess for this feature to really make sense to me ‘trying hard really with all my heart and soul :)’, maybe if it’s used as a real time editing of a level for a game like minecraft, maybe then i can finally rest and say, that could be a feature for a game. But from the looks of it, it is just being asked or meant to be used as a traditional poly modeling tool inside the engine.

I’ll keep trying… maybe one day… one day…

Unfortunately all the same to me at this stage :), I did check them out.

I think all i’m saying is that while its a good thing to have as a plus, it’s arguably not the most thing to be after with such passionate appeal. Especially (not to bring same argument back) can be done and tweaked and designed (as it normally happens in indie small and large studios alike) outside the engine of choice and into the designer’s plate on a 3d or even 2d app. everything else comes in as bells and whistles.

Last but not least, I wouldn’t have had any issue with this whatsoever had it not been the amount of threads i’m seeing related to the matter vs some threads that are far far more important to the engine in terms of gameplay tools that are still either missing entirely or half done.

So I just keep wondering about where the devs priorities lie these days in terms of development (game wise, not cars). Do they really wish Epic to spend time on such a feature with their limited time and resources vs other aspects? when you can do the same and more in blender and just add 20 seconds to importing a level. Are the devs imaginations or pre planning so low these days that they can’t visualize the level design unless it is in engine at all times? Is this what is absolutely going to make the workflow that much better or faster?

Finally yes i must say that some priorities are more important than others and it’s not a matter of just perspective. Example:

If modeling in engine tools (that can be done elsewhere for the time) are going to take X amount of time and delay Y development tools (that can’t be done anywhere except by hiring specialized developers and spending tens of thousands for at least a year, if you are at it with a small team), then it’s a no brainer i chose Y development by Epic that’s going to save me and others tons of time and money that none of us can do on our own within a reasonable budget and time and which is absolutely vital to a game.

Then again this is a tool probably being developed by an outside party, so good for them. only complaining and asking it to be production ready and integrated is asking Epic to step in when many of us wish Epic to do other things right now. We wish. Not demand. :slight_smile: