Hi man
For now I put aside the modular kit to build buildings (I don't want to go bald) and start building a dungeon kit
I wanted to ask you .... do you think it is better to have a single model with a unique texture or an exploded model to be composed in GG?
From the point of view of 3D modeling I definitely save a lot of time .... but then in GG? Do I need a lot of time and especially hair?
Forgive my bad english
Stay safe and have a nice day with GG

I think a single model is the way to go in order to have better performance and less drawcalls.

depends how big the model is, reasonable rooms is better than a massive object so you don't have issues with lights and culling etc
if you want full control and variation smaller is better - but you can always cut up sections later anyway

Both points are valid … just a matter of choice or intention really

I believe, and perhaps I am wrong, but in a model made up of several pieces, GG would treat each piece as an independent object, therefore it would make its own calculations for each piece.
I believe that this behavior would affect performance more negatively than a model made up of a single piece.

In addition, it not only represents more work for the artist but also for the end user, who will have to assemble all the pieces, after finding out where to put each one, and if the same thing happens to me with ikea furniture, I always screws left over. LOL

However it also has its advantages, the user will be able to use the pieces individually and only those that he needs in a certain scene, this system is much more flexible for the user.

I usually go in between, If its a large layout then maybe 3 sections so you can have more dynamic lights in each section without the hitting the any real notable performance loss.

Bold bit, not really, with grid snapping in place, you can add several layers without issue, without much issue and effort and have a really good complex level layout.Performance hit is minor at best. There is only so much you can fit on a 4k texture, and have to strike a balance between texture quality and diversity.To get any thing worthwhile working in gameguru a layered entity approach is the best solution, uses a lot less memory and resources then having a single mesh with multi textures, it really chows resources. The real benefit from my approach which I have been using for years you can have a single texture for the main level mesh, which drastically reduces resources needed, it also allows seamless shaders effects between main level meshes.

If you know how the grid system works a mesh layered approach can really make a difference to the overall quality and complexity of a level with a minor performance hit at best. A layered approach also allows you to overcome limitations imposed by DBO and X files formats.

Have to agree with wizard, if you know how to model using a single texture set for a large number of entities, and can get your head around modeling to work on the grid system it's far more optimized than larger models with their own textures or multi texture models.

Thank you all guys
I will reflect on your answers

Nice videos guys.
I also appreciate the advantages of working with multi-pieces, although I don't usually do it, I always think that it is easier for the user to put a house with all the furniture inside, than to have to put everything in its place, but as I said before this way it is less flexible for the user. The multi-piece system gives the user more freedom, for example parts to mount a tunnel, the user can make the tunnel as big as he wants.

Maybe having the opportunity to group all those pieces once assembled, in a single model in GG would help a lot.