What A $1,000,000 Game Looks Like In Unity (w/ Julian Ball)

This is this is actually our intro cut scene. So I might skip through this. Um I don't want to
How do you do the cut scenes?
Um I can actually just quickly show it. It's just like using the the timeline Unity timeline system and animations. Um, so it is just like uh I don't have the timeline open right now, but it's literally just a canvas uh with a whole bunch of yeah, these frames and then some text and then they're just
going from one to the other and just animating.
Um, relatively simple. They're just using some simple components just to make each one trigger the next one.
Sweet, dude. Yeah. Yeah, the way I do these is I do them in Premiere um and just do export a full video, not the text cuz it has to be localized. But for me, the reason I use Premiere is cuz I want full control over the the sound and everything to make it more cinematic. Um but my game is more like Bioshock, so I want it to feel more way more cinematic, like like an actual film. So this is a great solution, though. I love this.

Yeah. Yeah, it worked. But it sort of just I had not much time, so I quickly came up with it. Um, this is like another little little entry to the game. So,
bro, check out that water. Is that an asset? What? How did you do that?
No, no, it's all our own shader. Yeah. So, all the Yeah, pretty much. Like, we don't really use any shaders by other people. We just sort of make everything ourselves. The water's one of the big we spent a long time working on the water systems.
Can I see this in a scene view? Uh, Julian, can you pause it and we can look at the scene? Well, actually um uh well actually shown in the official Unity documentation for how to make good-looking water. And the funny thing is it was for URP and we do not use URP. They thought our water was so good that they thought it was URP when it's not.
Well, that's what's so funny is everybody thinks that URP by default makes your game look better and it's like no, it's just there's different reasons why people URP use URP or HDRP. It's primarily related to like a function that you like one little thing that you need is in URP and so you I

guess we got to use URP, you know.
Yeah.
It's not cuz it just by default makes it look better, you know.
So I'm curious about that mist that's blowing off of the boat.
Yeah,
that that is what looks really good. What is that?
Um just a whole bunch of particle effects. Like for example, this is all so so just like gross. Let me find, for example, there is a water splash particle here. And then I just have like an animation trigger. So like every time the boat comes and then sort of hits there, it just replays this this particle effect. And then I have all these particle effects going along the way. But that's like the main mist. And then this is just a scrolling like this is just an adaptation of our water shader on a plane that just scrolls. So basically has some noise and just scrolls.
The boat doesn't move. Everything's stationary and it just scrolls and I just animated everything just to sort of line up with how the waves just sort of match.
Yeah.
And um and then I just basically have

these big particle zones that sort of match the waves.
Oh my word.
Um so this is like a wave spray and then it just literally just has some smoke, dude.
It's a magic trick.
Yeah. All of it's completely smoking. You can also see like a depth shader that we have here to sort of mask out the water
from the boat so it doesn't go over it and it's just a bunch of cubes just placed.
Have you Have you checked out how this runs on like Steam Deck?
Yeah. Yeah. Yeah. So like we build for like it's available on Mac and also Steam Deck. We don't like officially support Steam Deck, but it runs I have a Steam Deck just here and it runs runs fine on on both um like pretty low devices. It's not a not a problem. Yeah. Like most of the time you get 60. it can dip a little bit below sometimes, but
um Steam Deck and um consoles are sort of like a a post 1.0 priority for us. It's it's not a Lens Island isn't launching 1.0 on console at all, but it is playable on Steam Deck and it there is controller support in the game, but it's we still want to improve it. We

still want to sort of refine it over time. We didn't want to rush to get Lens Island out on every single device and every format all at once and then have that take up all our development time and then the the underlying game be worse because of it.
Exactly. Exactly. What do we got here? A frozen uh
a frozen loading screen cuz Unity can't handle loading scenes [laughter] properly um in the editor.
Um but uh we're actually redoing all the splash art for Lens Island which is we're really excited to showcase the new splash art soon. So, this is the current sort of main artwork we sort of have for the game.
I will say this and I'm glad you said that. In my mind, when I saw that, I thought that doesn't look as good as the previous scene.
Um, there there it there's it it needs a little bit of a glow up and and so I think you're going to you made the right call there.
Yeah. Yeah. I think a sense of like uh depth and action and perspective and also multiplayer is is going to make a big difference um to the artwork in general. This looks so beautiful, dude.

So, this isn't isometric but or um orthographic, but it's it's it feels like it, right?
Yeah, it sort of has like um it it's like the top down camera angle, but I like to sort of think it's like an an evolved version of the top down.
There's some depth here. Yeah,
there is a bit of perspective. There's a bit of depth. There is camera rotation and there's camera zooming. You can zoom out and also sort of tilt the camera to sort of see in the background and then also zoom right in. Let's see what it looks like in scene view here. If you want to pause it or something. [snorts] Um, and let's see what it looks like without lighting and post-processing. If you could turn the post-processing off.
Yeah, I'll just pause it as well. So, here we go. So, like I said, here's our sort of chunk system. So, you could see that these are sort of the edges of the world. So, this is just one sort of portion of an island that's loaded in.
And then, uh, right now we're sort of standing,
I know, somewhere in the middle here. It looks good, but it's it it's encouraging to people watching probably that they're like, "Oh, this is actually what it looks like." You know, it's lighting

that does all the magic and and
um the post-processing and all that, you know, the fog and
Yeah. the lighting, even just the color grading does so much, dude.
Like the the whole time,
like from from the first day of making Lens Island, the color grading has done the biggest impact on the looks of the game. It's just uh Let's take a look at all of your different post-processing if you don't mind.
Yeah, I'll um I've got a few. So, we sort of have two we're also using the old post-processing stack, but we have sort of two layers. We have like
a layer of post-processing which does like in-game stuff. For example, like if I um
if I get like hurt, you know, like we have like, you know, like all these sorts of effects and we have
I use I sort of I do everything in a slightly sort of interesting uh not maybe non-advised way. So, for example, right, let's have a look at this like fire. Um, even I'm just set the time. Uh, okay. Set time and let's set it to like

actually a bit later.
That is so cool, dude.
So,
right now it's sort of at night time. And then you can see I've got this little sort of fire here. So, if I go to like our main camera, we have a um we use like a package for uh glow called MK Glow. So, you can see what I sort of do is like a combination. and I have this sort of strong intense glow that has a pretty tight sort of
um blur size. Um but then there's also a second glow that I do at second bloom which is actually like unity post-processing blue and that has a lower threshold but like a much bigger blur size. what it sort of ends up to think that there's sort of like light scattering where it's like there is this like little bit of bloom internally of the thing and then there's like large amount of bloom that sort of affects like the whole the whole area and it's just two blooms just overlaid on top of each other. So,
and I've I've definitely encountered this problem where it's like I want

certain things to have a high intensity and then I want the rest of the game to have a soft bloom. the soft bloom would be justified by I guess like moisture in the air, you know, um creating that scattered light.
So, there is that problem and I never knew how to solve it. So, it looks like you
you kind of had to solve it with your own sort of band-aid approach, which I love it. It looks great.
We could probably even see Yeah. a bit more. The world has has loaded now. It's going to rendering two views at once. This is going to be interesting on the performance.
Is exactly.
We'll probably see the world change as we walk around. Julian, is the is the land terrain? Um,
yeah, all Unity terrains. Yeah, you can see some cells sort of unloading there slowly on the bottom as I as I sort of walk around.
Are they are they generated with code or is it uh chunks of different land that snap together that are pre-made?
Pre-made. Yeah. So, every single rock and stick and everything in Lens Island was handbuilt. Like all the terrains are

hand painted. We don't use terrain like generation. Like I I mucked around with some terrain generation tools like Giaire and stuff like that, but none of them really gave me the control or results that I sort of wanted. A lot of Lens Island is actually birthed from restraint. It's like for example, we have like an interesting inventory system where uh everything is actually pre-laid out in sort of tabs and there is like item dragging and there are storage and stuff but it's sort of like everybody hates sifting through their inventory so let's just do the sifting for you and it's all there and that whole mechanic is just birthed from I didn't know how to make gridbased inventories when I first started making g games. It's sort of like it felt that like that's too hard. I I don't know how to do that. So, I'm going to make a different kind of inventory system. And um a lot of Lens Island is sort of built around what I would what was possible for me to make and then it sort of like

snowballed that over time. So, like a selling point for me is the fact that the world is handmade. And the reason why it's handmade is because it's like I don't know how to do procedural generation or sort of like make all these terrains and object placement via via code. And when I tried to do it, the results were just nowhere as good. um when I did it by hand. So then we sort of like met in the middle. I was like, "Cool, I'm going to hand build everything because I know how to do it. It's going to be more labor intensive. It take longer, but it'll give better results and we'll just basically procedurally organize all of these handbuilt aspects of the world."
So are these um are these AI uh animals are they using a nav mesh?
Yeah. Yeah. So there's a whole
Okay. So you're baking the nav mesh on the chunks at runtime. Gotcha.
Yeah. cuz we like Lens Island is, you know, a sandbox game, right? So, uh, okay, this is going to like completely ruin my performance, but it'll be good to good to show. So, if I go to

navigation tab, see, we can sort of see some some nav mesh build here.
And um, you can build a lens island anywhere in the whole world. So, you can build crazy castles and sort of do whatever. So if I sort of start building, we have to affect this nav mesh at runtime and rebuild it at runtime. So you can see as I'm placing these foundations, it's actually rebuilding the nav mesh every single time.
I mean, are you using nav mesh obstacles? Is that what it is? Or is it literally recomputing it without?
It's literally recmputing it. Yeah, cuz it cuz you know, nav mesh where I'm standing didn't exist before. Um we have obstacles in like the bottoms of the foundations to make sure that things don't go below [clears throat] them, but above here it has to sort of reconstruct it.
So um yeah.
Okay. Are you using inclusion culling at all?
I I can't imagine you are because it's all real time
to to an extent. We actually built our own um sort of occlusion optimization system that we use in the game. Mhm.

Um so like a lot of assets sort of have this this system that we um sort of built basically our own version of like occlusion occlusion culling also distance culling. So things basically step down the shader enable disable the shadows and do all sorts of like LODs based on distance to the camera and and sort of everything else.
Okay. So let's talk about your weather system or your dayight cycle and all that. Uh can you give me a breakdown of how this works? Um, I can I'll make a guess here that the directional light is not enough. Like you have to do a lot of post-processing adjustments. You have to tweak the color grading. You have to tweak the shadow colors. Is this true?
Yeah, definitely. It's definitely like a unison sort of everything. Um, so right now I just I just changed the weather to to rainy and you can see there are it's it's sort of there's like this slow it's sort of like light rain at the moment and weather sort of slowly morphs over

time.
Um, there's also like a a storm set.
Can you show me in the scene view? Can you show me in the scene view what's happening here?
So obviously a bunch of post-processing happening, bunch of particles. It's basically like a uh all the cell objects of the of the world here. So, I can close that.
I'm I'm I'm shocked that this runs well on uh Steam Deck to be completely honest with you.
There's so much going on.
A lot of Yeah, we've just we've had to solve a lot of creative problems of optimization and stuff over time.
Um but yeah. Yeah, I've I've played through almost the whole game on the Steam Deck and hasn't really skipped a beat. Um
can you play through the whole game? There's a there's a finish point.
Yes. So that that's what we're bringing in 1.0. There will be a credit role, an official final boss, final dungeon, a finish point, but it is in its core still an open world sandbox. So it's like in order to get that end point, you can take 40 hours or 200 hours, you

know, like you can really play the game however you want. And there's a lot of there are theoretically more than one end points. There's a lot of different satisfying completions to Lens Island. You can have if you're a farmer, if you're a builder, if you're a fighter, sort of.
Lens Island is a very hard game to make as far as game design because the whole concept of Lens Island is like, I want to play Diablo, but when I'm bored of playing Diablo, I want to play Stardw Valley. I want to play Minecraft, [laughter] but when I'm bored of playing Minecraft, I want to play this other game. And it's like, how can I get all these games to be under this one roof and cohesive? And in order to do that, you have to build four games worth of content. You can't just have one game that has like a slither of that amount of content. You have to just farm and play a farming game and it holds up on its own two feet and then it's like, "Hey, I'm I'm tired of my farming game. I'm going to do a bit of dungeon crawling and exploring and build a boat and find islands and stuff." And uh
it's been very interesting to make it all work from a design perspective.

So, and you by the way, Julian, you tell me when you got to go. I know you're busy. Um
that's fine. We could I want to I want to see if you don't mind your solution for the dungeons.
Yeah.
Um
yeah, we can we can look into one.
It's it's like okay making an exterior terrain based game just at its core just from an art perspective is really simple. Um the moment you go to interiors in in my game which is a first-person game everything changes. The whole the whole way the game is produced is completely different. Um, there's no terrain, there's no painting things.
It's building things out with ProBuilder. Um, and then set dressing. There's no painting objects into the scene. So, how does your interior dungeons work? Is it the same approach as the terrain?
Uh, the same approach that you just said where it's completely different to the above world, [laughter] but we also have sort of three uh three different world types in lens arm.

There's the above world, right, which is a procedurally generated sort of chunk based world, which is very very big, like very big.
Um, then there's a cave system, which is sort of a mirror of that above world, but underground. And it's a big sprawling procedurally generated cave network of mining and enemies and structures and all sorts of stuff. And that's also procedurally generated. And the two layers have to like talk to each other. And then there's dungeons. So then there's like dozens and dozens of these fullyfledged dungeons around the game that are these handbuilt dungeons that are basically just their own individual scenes and combat focused.
Yeah. And there's just like you're just straight up playing Path of Exile when you're in these dungeons except with a bunch of Minecraft and a lot more sort of sandbox elements also mixed in.
Um so I'm going to we'll open up into one of these. This is like the first tutorial uh dungeon right now.
And I'm just going to let us
So the underground is is still an explorative world and then a dungeon is combat focused with a maze that isn't procedurally generated. Is that what

you're saying?
Yes.
Yeah. So yeah, the cave system and above world are like handbuilt tiles using a grid system that are, you know, procedurally swapped around and then Yeah, the dungeons are just
a big scene and there's just no there's no real adherence to like a grid [clears throat] or chunk system or anything. And also just like the whole method of how we design the levels is completely different. Like the above we use terrains and everything in the above world. But the dungeons, as you'll see, they're sort of like floating in this void. And we very much went with this like minimalistic approach where when you just look at the scene and you just see this this sort of dungeon just floating there, you're like that looks a bit weird. But then when you're actually playing it in a game, it doesn't feel weird. like it just feels completely normal and you don't even second guess the sort of the situation that you're in. So, this is a sort of tutorial dungeon, so it's sort of a little bit cave here.
No enemies roaming around, no terrain. Um, some stuff is like optimized away. You can see some things disappeared, but you can see it is just this this simple level. And like it's just literally just

like the edge of the level. Like this is black in the scene. And that's like a significant part of
Lens Island. The level design is you can fall off the edge of the level and if you fall off you take damage and you respawn back on the level. So a lot of combat is very movement focused in Lens Island. And that's another thing. It's like it was a restriction. It's like I don't want to make this world where it's like, okay, a restriction to level design is I have to build like these huge cave walls on the side and have all this stuff down here. It's like that's so much effort. I'm going to have to build this whole camera fading system and all this other stuff. Let's do none of that. and instead it's a mechanic. And the mechanic is like positioning and your movement is very important in the game because you have these small little thin levels that you can you can fall out of and um and take damage.
So you can see just it's just a void and um and a lot of the time you're fighting enemies
on the edge of this sort of of this void. And
elements of the game,
light torches and stuff.

Oh my word, that's so cool. It looks like your textures are like all the same color and you're using lighting to create a dynamic color scheme. Um, is it true that they're all the same texture or is it just they're all the same hue? I guess what I'm asking is are these pre-built in Blender and then painted in like Substance and they all have their own texture map or are they all just a looping tiling texture?
Yes. So every Yeah, everything is the exact workflow you just said. So, we use Mayer. Everything is handbuilt in Mayer and then um and textured in Substance and they all have their own textures. But I definitely use lighting to do a lot of the heavy lifting. I'm looking at it like a black and white. It's like a black. It looks really I hope you don't take offense to this. It looks bad.
Looks Oh, it looks terrible. Like [laughter] you look at this and you're like, "Uh, okay." And then you look at that, you're like, "That's not the same game, right?"
No. That's why lighting lighting being having an understanding of cinematic lighting is so crucial with game

development. Yeah. Yeah. And the the other interesting like I guess optimization thing is um also a lot of these so for example right like this asset here this cliff
this is one single mesh that's a cliff and every single cliff you see in this whole level is just the same thing. It's just the the same cliff. I just rotated it and transcaled it and moved it around. But they're just all the same asset and there's like a couple little
variants. But to build like all of the rock structures in the whole game of Lens Island, there are dozens and dozens of dungeons. There's basically like three rocks and all the ground of the levels is this one ground piece that's just rotated. And it's like I just made things in ways where it's like, for example, uh if I have a cliff like that that's flat and then I merge into another cliff, I get I get Z fighting here, right? And like we don't want that. So then all the assets have a slight little bulge like this, like it's like a dome. So I can like make all the cliffs go inside of each other and mix and match them and all the all the sort

of textures line up and you just sort of never never really notice.
You know, I I'd love to see what it looks like without ambient occlusion as opposed because ambient occlusion solves a lot of the pro the problems of objects intersecting with each other and not looking natural.
We don't use ambient occlusion at all.
Wow. It just none of the game uses ambient occlusion in any way.
It looks so good. I'm wondering how you got away with that, honestly.
I can Let me find the uh I'll get into the post-process here cuz I guess I didn't properly show it before cuz it definitely is doing uh an insane amount of of of heavy lifting. Um, let me find the So, I think this one.
Here we go. There you go. That's no post-processing and with post-processing.
So, you can say the the colors quite drastically shift with the post-processing.
What exactly is what exactly is the

color grade you're doing? Are you adjusting the mid the mid and the highlights or what? Yeah, like a whole bunch of stuff. A lot of it is like the the basic sort of highlights. It's nothing crazy. Little bit of contrast, a little bit saturation. Um, but a lot of
You can remove this in the options, Julian.
Um, this color grading. No, this is just sort of baked into the game because it's sort of the way that I design the game is sort of interesting. I basically what you said before, everything looks like crap to begin with and then I assume that's going to go through this color grading pipeline and then it becomes what it's supposed to be and that even happens for the UI which is something that
wo
I'm happy with the outcome but it's a pain and you can see look like the whole UI gets the color grading part.
Okay, so the UI is not necessarily consistent across the game level. every single asset that I have to export, I have to export with a filter that basically I make I make an icon. I'm like, "Cool, that icon looks good." And then now I'm going to make it look like crap with a with a bunch of Photoshop filters [laughter]
and then put it into the game and then

revert it back to sort of how it how it should be. And the reason why I do this is just cohesion. like the the UI and the game world has the same color theory and the same rules and the same um tones applied and it just feels cohesive across the whole game no matter what you're doing.
Green in an icon is always sort of going to match the green of grass.
The gold for example in your UI has a real shiny bloom to it, you know, it looks like just really special
and that is because of that color grading, you know. Um, I love that, dude. Can you um can you and then we'll wrap up here, but can you talk to me about your color theory um in this scene? I'm seeing a lot of that high purpley red, high saturation purple red. Do you have intention behind your color theory or did you just wing it?
Yeah, that definitely intention like um I can also just quickly show you what it looks like in sort of other other times of day. It's going to take a little while for things to sort of load in. Um

but uh but yeah, you could sort of see a fair bit into the into the distance here. That's sort of another dungeon and stuff. Um but yeah, like color theory is always like very considered. Obviously, this needs to look this needs to look cold. Um but it also needs to look uh like a like a place that I can also sort of um come into I guess it's sort of like a bit of a mixed bag. The intention with this specific town was that it looks a bit bleak, a little bit a little bit cold, and almost a little bit uncomfortable. The color palette during the daytime, you can see it is a bit washed out, and
it doesn't feel super cozy, but it definitely looks pretty dramatic. Like it looks, you go, "Wo, look at this sort of town, the cherry blossoms. It has some
popping colors, but it doesn't necessarily feel homely." But then at the night time that really shifts and just like the huge amount of orange light cast everywhere um makes you feel

a lot more a lot more sort of nestled. Um, so like the the overall feeling that I sort of wanted to give was like this town is your safe space and
you you basically just see this this orange red out here and then everything here is blue and it's harsh and it's cold and you don't want to be out there. And it's just sort of really real sort of simple where like this town is like a bit of like a a beacon um where you can sort of see from distance and go, "Okay, up there maybe won't kill me. I'm going to go up there." Um, well, dude, thank you so much for sharing this. It is, I would say, the most complicated project I've looked at on this podcast and definitely intimidating for even me to see, you know, and I say even me just cuz I've been make I've been making games a long time and I'm really impressed with what you've done here. I'm just What? How did you get that water to move over there at the top right hand corner?
Oh, that's like a current. So, we have

the heck, dude. Yeah, we [laughter] have we have currents in our game and you can ride the currents on your boat and get little sort of speed boosts around the around the map.
Okay. Yeah, I'm officially jealous. Um [laughter] I'm not jealous of the eight years lost, but
this is great, dude.
Well, thanks for coming on that this is just amazing and and I I really I really am impressed and I know I can see here it's going to do well. It it's going to do well, man. I'm I'm really I'm feeling good about it and I could see why you're confident in it. So
yeah, like it's I have a lot of fun playing it and I think I've got a pretty good opinion of of what's a good game. So hopefully that opinion stays true. Um but yeah, like like I said, the the confidence can somewhat be uh
tangible because it's already released. Like we know people really like the game. For me, it's just very much been a journey of um can I just get to the end, you know, and be proud of it. For me, it's a bit of a difference between something that is good and something that I'm proud of. I think probably the

difference is knowing that I gave it my all. Like, I can make something good, but I could also at the at the same time know that I didn't give it my all and I I could have made it great.
So, I'm at this point where there's no doubt left in my mind anymore. I gave it my all. I think it's great. And I I I think other people are going to think it's great, too.
So, we'll see. We'll see how it goes. 4 weeks from now, the big release.
All right. Well, link is in the description to either wish list or play. You know, most people probably watch this when it's available to play. So, please everybody click the link in the description and um support this game. It looks amazing. Anything else in the description you wanted, Julian?
No. No. We just released um the release date trailer as well. So, if you want to see a little bit more about the game, you can like see the trailer. Is that on the Steam page?
Yeah, that's on the Steam page. We I I can't really
share too much, but there may be a way you can play the game
without paying for it at some point,
and maybe. So, maybe just keep an eye on if you want a little try of Lens Island.

Maybe something will come up in the future.
Brilliant. All right, dude. Thank you so much. That was a lot of fun.
Thanks for having me. Pleasure.
Yep. Cheers.