Vapourware Codexian Game Development Thread

[Nathaniel3W]

Hey guys, I'm wondering if you have any experience with stuff like this. In Unreal Engine 3 I have this level that has some farms. The farms have interactive foliage actors, which are basically decorative plants that you push around when you walk through them. Individually they are very cheap to draw. But here I have about 2000 of them on screen and my FPS drops from 60 to 40 when I'm on the farm. My rig is a 12-gen i7 with an RTX 3060.

In Unreal Engine 3, foliage is very very cheap to draw because it's all drawn as one instance. If they don't have collision, and they're low-poly like the grass, then they're almost free. Static meshes are a little more expensive. Interactive foliage actors are just a little more expensive than that. The vegetable field isn't bad. You can walk through that and push the plants around. I have levels with tons of interactive foliage like that and it renders just fine. The wheat is really costly though because of how many individual interactive actors are there.

Can you tell me some games that you think did wheat fields well? Are there any low-poly cartoony wheat fields you can show me? Were they interactive or purely decorative? I'm thinking I might have to replace all of my fun interactive wheat with regular non-interactive foliage. Do you know of any tricks to get the cost down?

Edit: Here it is in action.

A low-res gif doesn't really do it justice. It really is kind of a cute, fun effect. Maybe it's too subtle though. I'm going to see if I can make the mesh bigger, incorporating more stalks, so I can cover the same area with few meshes and thus fewer draw calls.

 

[The_Sloth_Sleeps]

Hey guys, I'm wondering if you have any experience with stuff like this. In Unreal Engine 3 I have this level that has some farms. The farms have interactive foliage actors, which are basically decorative plants that you push around when you walk through them. Individually they are very cheap to draw. But here I have about 2000 of them on screen and my FPS drops from 60 to 40 when I'm on the farm. My rig is a 12-gen i7 with an RTX 3060.

View attachment 48927


In Unreal Engine 3, foliage is very very cheap to draw because it's all drawn as one instance. If they don't have collision, and they're low-poly like the grass, then they're almost free. Static meshes are a little more expensive. Interactive foliage actors are just a little more expensive than that. The vegetable field isn't bad. You can walk through that and push the plants around. I have levels with tons of interactive foliage like that and it renders just fine. The wheat is really costly though because of how many individual interactive actors are there.

Can you tell me some games that you think did wheat fields well? Are there any low-poly cartoony wheat fields you can show me? Were they interactive or purely decorative? I'm thinking I might have to replace all of my fun interactive wheat with regular non-interactive foliage. Do you know of any tricks to get the cost down?

Edit: Here it is in action.

A low-res gif doesn't really do it justice. It really is kind of a cute, fun effect. Maybe it's too subtle though. I'm going to see if I can make the mesh bigger, incorporating more stalks, so I can cover the same area with few meshes and thus fewer draw calls.

What does it look like with quarter density?

 

[JoacoN]

The mollusk is here.
Finally finished Act 1 of my game after almost 2.5 years, however acts 2 and 3 are going to release much faster.
Year 1 of the project was figuring out how the engine works, then spending almost another year making a full prototype with all the mechanics for the game and developing a style and story. So the game as an actual playable thing only really started existing between last year's October-December.
Anyways, most important stuff is the first act is done, the second boss was added, and there is a new weapon.

Changelog

[New]:
- Finished Act 1 of the game.
- Added the second boss of the game, Chigrepe, a strange ancient creature that protects the second Crystal Fragment.
- Added the Chripete, a new weapon dropped by Chigrepe, stab enemies with this mini tongue to deal a lot of damage, and chain together combos to keep yourself in the air!
- Cyrstal Fragments now emit ominous music.
- Added the Chigrepes logo to the game, representing the game's dev team

[Changes]:
- Reduced the health of the first boss from 3250 to 2665 to balance the difficulty curve.
- Added "boomer" to the special thanks section on the credits.
- Adjusted some text of the footbal match quest.
- Increased the health of Zeums.
- Replaced the look of Crystal Fragments on the HUD.
- Reduced the volume of enemies teleporting.
- Reduced the volume of low health music.
- Made boss arenas and portals be hidden after beating them to improve performance and their areas not re-enterable.

[Bug Fixes]:
- Optimised the bosses health bar.
- Improved performance on HD Mode.
- Fixed some NPCs rendering on top of some models.
- Fixed painting the same space for a little over 10 seconds causing massive lag spikes.
- Fixed a bug were the low health music would play over the pause menu.
- Fixed the low health UI not being hidden when pressing F1.
- Fixed being able to talk to El Buhonero from inside the vents.
- Fixed the Chocolatada contents clipping through the model at certain angles.
- Fixed some issues in the sports are and fixed 2 missing materials.
- Did minor adjustments to the pause menu.

 

[Nathaniel3W]

The_Sloth_Sleeps Thanks for taking a look.

Here it is at half density. I suppose it still kind of looks like a wheat field. But it's definitely sparser. Maybe I can get away with it. Maybe with a combination of more stalks per mesh and lower density I can get the performance up.

 

[Nathaniel3W]

The mollusk is here.
Finally finished Act 1 of my game after almost 2.5 years, however acts 2 and 3 are going to release much faster.
Year 1 of the project was figuring out how the engine works, then spending almost another year making a full prototype with all the mechanics for the game and developing a style and story. So the game as an actual playable thing only really started existing between last year's October-December.
Anyways, most important stuff is the first act is done, the second boss was added, and there is a new weapon.

Congrats on the progress. It looks very Prosper-like. Getting the first instance of anything is the hardest. Now that you've made act 1, act 2 should go a lot faster.

 

[JoacoN]

Congrats on the progress. It looks very Prosper-like. Getting the first instance of anything is the hardest. Now that you've made act 1, act 2 should go a lot faster.

Ye act 2 is fully planned and I have most of act 3 planned too
+ 2 friends are helping me by doing the music and some sound effects, so now it's like a sort of small official team, which is why the game had a little shuffling around and this update took some time.
Only thing left for this is a small patch with some extra stuff I wanted to release late this month or early next month, and then I can start act 2 with 0.7

 

[The_Sloth_Sleeps]

The_Sloth_Sleeps Thanks for taking a look.

View attachment 48934

Here it is at half density. I suppose it still kind of looks like a wheat field. But it's definitely sparser. Maybe I can get away with it. Maybe with a combination of more stalks per mesh and lower density I can get the performance up.

Yeah the first thing I usually do is look at how to draw it simpler before looking at optimizing. How does it look with larger stalks to compensate for the fewer amount? Also maybe also you don't have to space them so tight. And maybe ones in the middle don't need so many.

 

[RobotSquirrel]

1. You give a big portion of yourself to a project and then see streamers go "what a shit game man haha I'm uninstalling" - it's soul crushing.
I've heard people use the expression, but I never experienced it before. And it's not even your fault, and your couldn't have done anything else. The programming part you delivered did its job.

The easiest way to fix this problem is to not use the success metric of being popular and liked. Instead realize you were able to do something that the majority of people - especially streamers can't do and take pride in the fact that you shipped and that your game didn't die in development. You may not get the monetary success on this attempt, but you definitely got the success of completing the project. That's how we need to re frame things. Because the financial success metric on average goes to prove there is not enough business out there for the average game developer, steam is covered in shovel wear indie shit and streamers are trying to take the place of games journalism to curate and gatekeep success so that their friends are the only people that survive and become successes.

Plus in business, unless the person has a specific want for your product you're trying to sell to someone that most likely doesn't want your product which is why you're going to get negative feedback. Customer motivation from what I've seen is the biggest failing point most developers succumb to, simple fact is people don't want your game and lack the incentive to want to go out and even try the game, they have the comfort of their bloated back catalogue, they have numerous popular F2P games now, so if you're trying to sell a product against those forces that is going to be the reaction. So if you're an indie and you're trying to charge an upfront cost of $20 you're setting yourself up to fail because no one is going to buy your game and if you provided a demo they have even less incentive to buy your game, its probably better to release prototypes as complete games and sell a more refined version of the idea in the same way Dwarf Fortress did because you need people to like your work before you can sell it to them and you can't do that unless they've played it.

Finding someone to buy the game is the challenge and building a community to evangelize you as a developer should be the goal, you don't want a community of influences you want a community of fans and slowly build that up over time, this is the mistake people keep making, they just want to become viral. That shit doesn't happen, if you look at the history of some of the most successful indies, their catalogues reach into the double digits before they got a hit. There are also unknown factors like the person being an industry veteran and having connections with publishers and certain organizations - that certainly helps a lot. Not everyone gets to be that though.

But if you really want the ego boost, I recommend showing your game off at a indie showcase because 99% of the people who play your game will be awesome and very supportive. It helps make the horrible imposter syndrome feelings go away, especially when you get a kid playing your game and they think you're cool because you make video games. I'll say don't let it discourage you, if you keep going with the hobby of making games it is rewarding but you have to get your mind out of seeing the success metrics the same way that everyone else does because its that survivorship bias. You will get there, just keep going.

 

[Nathaniel3W]

Wow. The most positive and encouraging thing I'm going to see today came from RobotSquirrel on RPG Codex. Most people here would have told Bester "lol you suck."

 

[what am i doing]

Hey guys, I'm wondering if you have any experience with stuff like this. In Unreal Engine 3 I have this level that has some farms. The farms have interactive foliage actors, which are basically decorative plants that you push around when you walk through them. Individually they are very cheap to draw. But here I have about 2000 of them on screen and my FPS drops from 60 to 40 when I'm on the farm. My rig is a 12-gen i7 with an RTX 3060.

View attachment 48927


In Unreal Engine 3, foliage is very very cheap to draw because it's all drawn as one instance. If they don't have collision, and they're low-poly like the grass, then they're almost free. Static meshes are a little more expensive. Interactive foliage actors are just a little more expensive than that. The vegetable field isn't bad. You can walk through that and push the plants around. I have levels with tons of interactive foliage like that and it renders just fine. The wheat is really costly though because of how many individual interactive actors are there.

Can you tell me some games that you think did wheat fields well? Are there any low-poly cartoony wheat fields you can show me? Were they interactive or purely decorative? I'm thinking I might have to replace all of my fun interactive wheat with regular non-interactive foliage. Do you know of any tricks to get the cost down?

Edit: Here it is in action.

A low-res gif doesn't really do it justice. It really is kind of a cute, fun effect. Maybe it's too subtle though. I'm going to see if I can make the mesh bigger, incorporating more stalks, so I can cover the same area with few meshes and thus fewer draw calls.

If you're able to give them the same appearance (interactive when fully at rest vs instanced), could you swap an instanced piece of foliage with an interactive one only when the player gets close? That way most of them would be instanced, and the more expensive interactive ones would only be a minority.
Alternately, what about making most of them instanced and only a smaller quantity (1 out of every 10?) interactive? This might give a very similar effect when walking through (would the players notice that not all of them are moving? perhaps not) but allow you to retain a dense field.

I do think the half density looks fine. You could help reduce the sparse appearance by giving the ground under them a more pale brown, earthy appearance (something similar to the wheat itself) instead of the current one (which, I must say, looks like gravel/rocks - looks fine on the paths, immediately leapt out to me as jarring on the fields). When thinking about other games that had low-poly but decent-looking crop fields Guild Wars comes to mind, and while their crop fields were mostly somewhat sparse, they generally had a nice earthy tilled soil appearance that helped sell it as a crop field, even if the tilling pattern sometimes didn't match up with the kind of crops being grown.

 

[Nathaniel3W]

Still working on the wheat. Changing the texture under the crops should be really easy. I can probably just use the same texture but tile it smaller and multiply it by a different color. The texture density will be different (I'm trying to keep it really close to constant to help with the pixel aesthetic) but it probably won't be noticeable since it'll be under all of the wheat. I'll take a look at Guild Wars. Thanks for the suggestions what am i doing

Working on a forest area next to the farm area. I wanted a fog effect to sell the density of the impassable tree areas. I'm pretty pleased with how it looks. I ran into a complication, but I think I can work with it.

UE3 is really picky about fog volumes. You can't modify their shape except for translation and scale. So what I did was make a second landscape--which UE will let you modify, raising parts here, lowering parts there--and instead of a "landscapy" material, I put a fog material on it, and then raised the fog where there are trees and lowered it below ground where there are no trees. Now I can put fog exactly where I want it in the shape I want. Great. But UE3 doesn't let you turn off collision on the landscape. So I'm going to have to be really careful to make sure all of the fog is well behind the tree line.

Edit: By the way, the lighting in these two areas is pretty much the same. I'm using different color grading lookup tables in the two zones to get a different look. The farms feel sunnier with more yellows and the brights pushed up closer to white. The forest has bluer darks, which I think helps make even the sunny areas feel shaded.

 

[Zanzoken]

I have finally started learning Blender. My goal is not to be able to become a competent 3D modeler and make assets from scratch, since it would take years to reach even a moderately acceptable level of skill. Rather, I want to be able to take store assets and tweak them to suit my tastes and have cohesive art direction.

So the model on the far right is from a Unity asset pack. The three models on the left are my attempt to deconstruct and modify it.

The biggest issue was when I imported the model into Blender, I found it had a ton of excess geometry under the surface that made it difficult to edit. I really don't understand what the modeler was doing there. I didn't take a screenshot of it with all the pieces I had to pull apart, but the wireframe gives an idea. You can see how many fewer edges there are in my edited version (second from right) vs the store version.

The changes are subtle but I removed the extrusions from the cap and crown of the helm so that the helm is smooth, and removed the split down the middle of the visor. I will probably make other tweaks but overall I am happy with the results. The fact that I was able to separate the helmet from the visor means I can have an open face version, which is a nice bonus.

Open to any tips or feedback that anyone may have!

 

[RPK]

I have finally started learning Blender. My goal is not to be able to become a competent 3D modeler and make assets from scratch, since it would take years to reach even a moderately acceptable level of skill. Rather, I want to be able to take store assets and tweak them to suit my tastes and have cohesive art direction.


Open to any tips or feedback that anyone may have!

About a year or so into my gamedev journey, I decided it was time to start learning blender. I quickly came to the same conclusion about years and competency. It seems like your goal is realistic. I'm curious to see how it goes for you.

 

[RobotSquirrel]

Open to any tips or feedback that anyone may have!

use floating intersecting Geometry rather than extrude everything, it lowers tri counts by a lot, reduces the complexity of the work and UVing becomes a lot easier. Plus it makes the mesh look less bumpy. Look out for wasted edges, if your edge flow isn't changing much those edges can probably be merged.
Also smooth your normals as it makes the geometry look higher poly.

 

[bionicman]

Zanzoken what kind of assets will you be making? I do game dev on the side (although I technically work on games as part of my real job too), and would like to become a full time indie one day, so I have a habit of collecting unity assets like pokemon. The kind of assets I would find the most useful are synty-like assets, since i mostly use synty in my project, but for things synty has not covered. E.g. monsters in their style which are not humanoid or even some animals... I could not for a life of me find a 3d model of a low poly rat in that style, so I had to make one myself, and rat is the most common rpg enemy critter. Anyways, I think if you tried to fill in the gap of synty, everything they haven't done, you'd probably make a lot of $$$.

 

[bionicman]

A game I programmed was released this year. I was only a lead programmer, the game director was the guy with the money and he wanted things done his way or the high way. Some of the game design wasn't good, I debated it with him until blue in the face, then eventually he couldn't take it anymore and became kinda hostile to any feedback I gave, so I stopped giving it. I managed to convince him to make some changes at the beginning, but it was a drop in the ocean.

Two things to say now that it's released.

1. You give a big portion of yourself to a project and then see streamers go "what a shit game man haha I'm uninstalling" - it's soul crushing.
I've heard people use the expression, but I never experienced it before. And it's not even your fault, and your couldn't have done anything else. The programming part you delivered did its job.

2. Whenever the design doc was lacking in details, I filled in the gaps with my own ideas of how things should be. There's definitely a portion of my game design in it.
Example: the streamers (I watched a lot of them for days) could never figure out how to split the stack in the inventory. They ended up convinced that it's unsplittable. It was terrible seeing this.
I did it like in World of Warcraft, but apparently that didn't translate at all. If we had playtesters, I would've seen it and changed it, but we were on a tight budget.
Looking at these design decisions (that are mine) falling flat on their face, big time, was another soul crushing experience. Not even humbling, but soul crushing.

To have a game you worked on super hard come out to "REVIEWS: NEGATIVE" on Steam is one of the saddest things you can get in any profession. I didn't understand how people could get emotional about their games before or talk so much about them 20 years later like Tim Cain does. Now I see.

idk if you ever worked at a real game company, but I've had the same experience dealing with designers and directors. They will make dumb decisions and be stubborn about it. Whenever I get some sort of request I question it to death, i follow this muskie advice:


you shouldn't look at this experience as something that stops you from trying again, you've learned a lot from this (probably). Yes, maybe it was a soul crushing experience, but for me, all the soul crushing moments in my life made me a stronger and smarter person (after all the girls have rejected me, i'm still hopeful ), I hope you see it this way too.

 

[Zanzoken]

Zanzoken what kind of assets will you be making? I do game dev on the side (although I technically work on games as part of my real job too), and would like to become a full time indie one day, so I have a habit of collecting unity assets like pokemon. The kind of assets I would find the most useful are synty-like assets, since i mostly use synty in my project, but for things synty has not covered. E.g. monsters in their style which are not humanoid or even some animals... I could not for a life of me find a 3d model of a low poly rat in that style, so I had to make one myself, and rat is the most common rpg enemy critter. Anyways, I think if you tried to fill in the gap of synty, everything they haven't done, you'd probably make a lot of $$$.

Thanks, I'm not trying to be an asset creator but rather get into indie dev myself. The game I want to make is gladiator-themed. I already found a character model I like for the gladiators, so I will mostly be focusing on creating customization options (i.e. faces, hairstyles, etc) and then weapons and armor. The art style I'm going for is low-poly and stylized, but not quite as abstract as Synty. I did buy their Rome asset pack though and intend on using their environment assets to build my levels, though I will modify them in Blender first.

 

[Justinian]

The_Sloth_Sleeps Thanks for taking a look.

View attachment 48934

Here it is at half density. I suppose it still kind of looks like a wheat field. But it's definitely sparser. Maybe I can get away with it. Maybe with a combination of more stalks per mesh and lower density I can get the performance up.

Have not worked with 3d so take it with a grain of sand but would it be cheaper resource wise to just have them move according to a pattern (so that a wave is caused or something) when you're moving near them rather than have actual collision?

 

[Nathaniel3W]

The_Sloth_Sleeps Thanks for taking a look.

View attachment 48934

Here it is at half density. I suppose it still kind of looks like a wheat field. But it's definitely sparser. Maybe I can get away with it. Maybe with a combination of more stalks per mesh and lower density I can get the performance up.

Have not worked with 3d so take it with a grain of sand but would it be cheaper resource wise to just have them move according to a pattern (so that a wave is caused or something) when you're moving near them rather than have actual collision?

No, probably not.

There are two things causing my slowdown: collision and draw calls.

Collision in a 3D game engine is done through an octree lookup. An octree lookup is fast. But if you have a lot of small objects close together, then the game engine could get bogged down as the player approaches. No matter what kind of actor I use to represent the wheat here, I will always have to query the game engine: Is the player nearby?

Draw calls get handled differently in different game engines and what kind of actor is getting drawn to the screen. In Unreal there are certain kinds of actors called instanced actors that can be drawn in one draw call. They just need to share the same mesh and the same material. The UE3 foliage system is one example of instancing, where you can have thousands of meshes on the screen, and they all get drawn in a single call. The GPU still has to draw the triangles, but it's still a lot faster than sending a draw call for each individual mesh.

I think it's impossible to draw interactive foliage as an instanced actor, all of them in the same draw call, because each individual actor has its own data in the vertex shader. Each one has to say individually whether the player is bumping into it.

So if I could implement your suggestion, I would still need to look up collision, and I would still need to have per-actor information (besides location and rotation) that would affect the draw, which would mean I couldn't draw them all in the same call.

I suppose I could try to do some trick to have it as regular non-interactive foliage until the player reaches some single larger external collision box, and then replace the non-interactive foliage with interactive foliage. That's basically still the SOP in UE5 so far as I know. (https://www.reddit.com/r/unrealengine/comments/1628r29/the_tutorial_ue5_over_300000_performant/)

I could implement a system like that linked above, but for now I think I will just need to reduce the density somehow, probably by making the mesh bigger and incorporating more wheat stalks per mesh.

 

[beardalaxy]

Nathaniel3W out of curiosity, is there a reason you're using Unreal 3 over 4 or 5? Not that there's anything wrong with it necessarily, I just am not familiar with why someone would want to do something like that since I don't use Unreal.

As for my own progress, I have decided to ditch the idea of changing the tileset. I've just designed so much of the game around the default assets and changing them to something else will be a lot more of a ballache than just changing them. Then, that's if they even look good in the end, which I have dubious thoughts about. I finished the voice acting, which means more or less I am in testing phase now, so I also am wanting to start getting the game out there and showing it off... so if there is all the sudden a big shift in graphics in a few months, I feel like that could spark a little bit of controversy. More than saying I'm not doing it at all, if people find that they would have preferred the game with the original assets. Thankfully, the RPG Maker default graphics are still really good and I think they blend well with the rest of my custom art as being backdrops for everything else going on.

Anyway, the game will be going into beta soon. I'm going to start doing a playthrough of it myself to try and find as many bugs as I can, then once I finish that I'll recruit a bunch of folks to help out. I'll be particularly looking for feedback on the battle and economy balance, since those are two things I have pretty much no experience in despite using RPG Maker for 18 years (I tried to stay away from battles as much as possible).

Since the look of the game is more or less finalized now, I've got some screenshots to share at long last! I am thinking of making a thread on here as well, but there isn't much in the way of interacting with the game or anything since there isn't a game download so I think I might wait until beta goes up to make a thread about it. Until then you can pretty much get all the info from Itch.io.

 

[Morpheus Kitami]

The_Sloth_Sleeps Thanks for taking a look.

View attachment 48934

Here it is at half density. I suppose it still kind of looks like a wheat field. But it's definitely sparser. Maybe I can get away with it. Maybe with a combination of more stalks per mesh and lower density I can get the performance up.

A thought, what happens if you make each individual wheat twice the size they currently are but still within the same amount of space?

 

[what am i doing]

I think it's impossible to draw interactive foliage as an instanced actor, all of them in the same draw call, because each individual actor has its own data in the vertex shader. Each one has to say individually whether the player is bumping into it.

If it isn't supported in UE3, that's not the reason why. Every instance has its own transform data, if for no other reason than to determine where in the world it appears. More likely the engine's batching process is just less sophisticated than UE4 or UE5's, being older.

While I'm not accustomed to complex animations (bones, multiple transform matrices corresponding to them, etc) I don't see a technical reason why these would conflict with instanced rendering; the only restriction is that whatever number of matrices per object your object with the highest number of matrices requires, all objects must use that number of matrices worth of space for their index in whatever buffer it's stored in, even if some of them don't use all the matrices. If all you're drawing in a given call is objects that all use the same mesh anyways, then that's not a problem, because they'd have the same number of bones and the same number of matrices in the first place.

Of course that doesn't rule out the CPU side workload of handling collision and transform adjusment. But it's not a fundamentally impossible task when rendering and I think if an RTS or similar wanted to have say a ton of soldiers on screen in different stages of their animations they would want to do it with instancing, if they cared about performance/optimization. But they would probably put... tanks, or whatever, in a separate call due to having different number of bones. A dedicated FPS engine might not bother, if they only expect a few dozen such actors on screen at a time. Low draw call count.

tl;dr engine limitation, not API/hardware limitation

 

[Nathaniel3W]

Nathaniel3W out of curiosity, is there a reason you're using Unreal 3 over 4 or 5? Not that there's anything wrong with it necessarily, I just am not familiar with why someone would want to do something like that since I don't use Unreal.

My sad story starts about 10 years ago when Unreal Engine 4 was in closed beta testing. It had started learning UDK and had made a little progress on my previous game Himeko Sutori. I was maybe a year into development when UE4 was made available. I didn't want to start over from scratch so I continued working on Himeko Sutori while the rest of the world moved on to UE4.

After many years I released Himeko Sutori. Then I started working on Septaroad Voyager as a mod to Himeko Sutori. I had seen Jeff Vogel's GDC talk about how Spiderweb Software runs a profit. He releases not-great games very quickly, reusing code and assets from prior games. Then every few years he makes a sequel or a remaster. I wanted to release my next game quickly, so I kept using UE3 because that's what I knew and that's what I had assets already loaded into. Hopefully if I can finish this game in the next 6 months to a year, then the total development time won't be too horrible, and maybe my income divided by time spent will be something above minimum wage.

I promise after this I am learning to use a modern engine. Or maybe I'll get out of game development entirely and go learn a useful skill.

A thought, what happens if you make each individual wheat twice the size they currently are but still within the same amount of space?

It would probably look OK if I doubled the width of every rectangle and fit two or three stalks on each. Then I'll add more rectangles per mesh, and hopefully it'll look just as full with fewer meshes. I'm going to get on that. Thanks for the suggestion.

what am i doing thanks for the info! That's really outside my specialty. I'm definitely more of a gameplay- and UI-programmer and not an engine programmer.