Making a tree
polycounter lvl 7
Making a nice high rez tree over the next few days. Figured I would post some of my work so far. Going for a high fidelity texture that holds up close, and uses tessellation and normal maps beneficially together rather than separately.
First few days was some visual tests and making the base texture. Going to start work on the actual mesh for the tree, and move to the leaves and twigs pretty soon. Going off of a black cherry tree for most of my reference
First few days was some visual tests and making the base texture. Going to start work on the actual mesh for the tree, and move to the leaves and twigs pretty soon. Going off of a black cherry tree for most of my reference
Replies
Thanks a ton! Yeah it is, kinda. I did a good deal of Nd02 experimenting with a flat image to see what worked and what didn't. I ended up using a slightly adjusted image (black and white high frequency looking heightmap thing) plus the original image and various passes to get the final result. Something I did specifically take into account was the heightmap for the tessellation. I wanted the normal to compliment it rather than try to achieve the same goal, so I ignored certain shapes when making the normal map.
Yup! This is just taken in the original marmoset. I pulled in their forest cover sky and then tweaked the DoF settings slightly to get what you see there. Also thanks cougar!
Hey man, thanks for the tip. I've been wanting to check out knald for some time ( I was actually going to use it on this till I got such a solid result with the heightmap I already had). Definitely a thing to try for next time though!
This weekend I spent my time trying to settle on the right approach to making the trunk. My options in my head, and conclusions were this (keep in mind this is for a base mesh, not something to bake a normal from):
Speedtree -> don't own, cant export from demo version, UDK only viable for now, that one is out
Hand model -> viable, but slow, unwrapped can suck
Zbrush -> doable, but fine branch work gets tricky and slow
Paint effects -> control is limited, but has good UV options
I spent my weekend trying out each option and decided to get a test mesh in from both zbrush, and from paint effects. The zbrush one turned out alright, but was not worth it's time at all for the final shape (something I could have hand modeled in no time flat).
Something of interest to note was that decimating a sculpt to put my texture on, did let me fiddle around with what differences geo levels made. I tried 2 meshes, one at 1k tris and another at 4k and noticed that because I was rendering with tessellation, the levels made no difference. For my paint effects attempt, I kept things super low (especially on the smaller branches) to see what kind results super low poly work would bring in.
Again, it was all tests. For the most part I wanted to learn about different workflows and experiment with finding something that worked. My final call will be to use paint effects. Honestly the main reason for this is converting paint effects to polygons in maya makes for a UV perfect tree (a HUGE time saver). Looking at what I had at the end, there were some things to take note of:
-4 divisions is just too small for medium branches
-My texture is effective at certain scales for covering up mesh intersection (a big worry of mine)
-Variety in model silhouette can hide lots of repeating texture-ness (a real world I swear)
-Tessellation breaks down a bit with a global scale on tiny branches. I may make a non tessellated material for those branches.
-UV scale can change a bit between branch sizes but the small branch sizes need a smaller UV scale in general to get rid of some of that ugly noise.
Anyway tomorrow night Ill make my final base mesh pretty quickly (probably keeping it as a paint effect so I can add leaves through that method later) and move forward from there (I'm thinking twigs and leaves coming soon).
Not this time. All done in marmoset. Could consider tessellation magic perhaps though
Anyway, its not perfect but I am pleased to get it as far as I did. Was a great study, but its time to move on to something completely different. I may make some separate renders with marmoset 2, since I am still using 1 and rand into certain shadow darkness colors when using a spotlight. Ill get some pics up later today of my textures and leaf cards.
Hey man those are pretty cool! Are you working from texture source or are you baking them down?
Alex
@ikarus, Thanks man!
Sorry for the slow response to this thread. I ended up taking a few weeks of for the holidays then got slammed right as I got back. Below are the pictures for those curious. Tri count for the entire tree came to around 30k.
Was a great project to learn a ton about pipeline. Here are some conclusions I came too for those curious:
-Good bark can cover up for a lot of problems with the mesh/UV seams of the actual tree
-While smaller leaf cards are nice, it is more efficient to make bigger cards with more leaves that you can later split into smaller ones. You could include a twig card in this.
-While some programs like speedtree can be very helpful with prototyping, hand placing /modeling gives unrivaled control. In my experiences here I would say hand placing will be what I do in the future to save myself time
-Testing can save tons of time, but also create lots of time. I saved myself hours with some quick R&D with the textures (testing normal maps before making stuff tile) but I created hours off work learning various tree pipelines when I have just stuck with common sense techniques in some areas.
-There are some definitive restrictions to rendering leaves within the original marmoset. The biggest I found was that there was no shadow color or intensity control. As a result when the transparency of the leaves caught the shadow, I would get super dark areas of my renders. Specifically where AO was also in effect. Unfortunately turning off either resulted in sub par renders.
For next time:
Bigger cards, more color variation in the tree
Lower tri count, better results
Not picking a black cherry tree, as their branches are just too weird (they fall in between different things/styles our brains are used to seeing
Also here are my reference shots for those curious!
As always feel free to PM with questions, etc. Also around on twitter (Dralex789)