# Revealing New Realistic Game From India?

## Метаданные

- **Канал:** Varun Mayya
- **YouTube:** https://www.youtube.com/watch?v=9Csd_qZyt3c
- **Дата:** 14.04.2025
- **Длительность:** 28:29
- **Просмотры:** 636,037

## Описание

8 Months, 40 People and after a monumental amount of hard work later...we have something to show.

## Содержание

### [0:00](https://www.youtube.com/watch?v=9Csd_qZyt3c) Segment 1 (00:00 - 05:00)

Heat. Heat. Heat. Heat. Heat. It's been a while. This entire thing that we're about to do right now, it's going to be short in one take. There's no script. Uh we're doing it a day in advance because it's just easier. It avoids all the tragedies that can occur with power outages. But ladies and gentlemen, in the last two years, AOS has really grown. We're close to about 300 employees at this point. And it's just been a phenomenal ride. We have set a benchmark when it comes to video, not just across India, but globally. And a lot of people in India now recognize that. globally recognize our standard, the standards we've set in video. But I wasn't satisfied. As we grew, I said, "There's something I've been wanting to do since I was a kid, and it's very expensive to do. It is technically very difficult to pull off. It requires too many things to come together. " And I said, "If I don't, who will? Someone's got to try. " So, I've hyped this project enough. Let's get straight into it. Ladies and gentlemen, today I want to introduce you to something that's taken 8 months, 40 people, really skilled people, and we've done it in utmost secrecy, and I hope you really like it. Ready? Here goes. I bet there are a lot of people leaving from the live stream right now. It's a watch. It tells the time. It's not smart. There's no AI. I mean, what more do you expect from creators or influencers or whatever?

### [5:00](https://www.youtube.com/watch?v=9Csd_qZyt3c&t=300s) Segment 2 (05:00 - 10:00)

Right. This is the best that uh us as a class can do. Right. I think we can do a lot better. In fact, this entire box here, it's misdirection. We don't actually need it. I practiced that for a bit. So ladies and gentlemen, without much further ado, I hope you're watching this in 4K because let the footage roll. I'm so sorry. Heat. Heat. Hey. Wow. Heat. Heat. Hey, What you saw was really special because it's just a small fraction of what's already built. And there's something very special that happened. A very specific technological shift that just it's not even a shift. It's just something that got really good. And no, it has nothing to do with AI. There was negligible AI in what you saw right now. And I don't expect text to 3D to be really good this year. I don't even expect it to be good next year. So this was done in an old school way but in a very different manner with a very different pipeline. And I want to shed some light on it because it was so much monumental effort and like I told you what you saw was just a fraction of what we have already built. You see we set a

### [10:00](https://www.youtube.com/watch?v=9Csd_qZyt3c&t=600s) Segment 3 (10:00 - 15:00)

benchmark in video content. Globally we're known for that. So we now want to set a benchmark in this new industry. And we are going to do it by setting our standards for quality so high that when you see it, you can't tell where it came from except for the where the story has come from and what the assets kind of look like, where the origin of the assets is. But I had somebody recently go out and put another very high-quality AAA game that just came out, some screenshots of it, compared to screenshots of some of our levels and some of the assets that we had generated and created without any of the set dressing because sometimes that can mislead people and without any of the extra environmental work and ask people to rate it. Most people rated us either on par or above the other competitor, which is crazy considering that this is of Indian origin and this hasn't been done so far. But I need to tell you the secret because it is a big one. 100% of the following screenshots that you're going to see on your screen right now were captured in engine. Take a look. So, what changed and why is this special? It's special because we now have one of the largest scan libraries for a game in the world. We have over a thousand assets. That's right. Over a thousand assets that's been scanned specifically for this game with the highest of quality cameras with technique that only a video company would really understand well. And it's very easy for us to adopt a photogometry technique, at least the way we're supposed to do it. it to get super high quality assets. We've adopted it. And the way we did it was crazy. We blocked out an entire town. That's right. We went out and blocked an entire town, the town of Chanderei. And we had our team scan it. a very expensive affair. I wouldn't recommend to anybody watching. But not only did we send up video teams there, we not only scanned the buildings, we also used drones to scan, you know, floor 2, floor 3, uh, 15th century architecture. It really was the most grueling process for the team. And that was followed up by the most monumental 3D cleanup effort that I think any team here has undertaken because just scanned assets are not good enough. You also need to be able to clean them. have an actual artist with 3D modeling, not just 3D modeling skill, but artistic sense to go in and say, "Well, the window sill is not perfect. I need to fix that. " Well, the handle of the auto is not perfect. Well, the front of the um, you know, uh, scooter is not perfect. I need to replace that, you know, area. Somebody with that skill and we had to have a huge team of them actually sitting together cleaning up all of this. Really, if I knew everything I knew both about the expense and the amount of coordination this takes, I probably wouldn't have done this. But thankfully, we're done now. And we have the largest scan library, one of the largest scan libraries for a game ever. I want to show you some of the scans. This is something that will hopefully interest you. As you can see, this is not a photo. It's not a picture. This is actually a 3D asset. And as you can see, even with things like the fruits, we had to actually go scan them. We have to make sure things that can't easily be scanned like tires and spokes. We had to remodel. So it's not this magical technique where you press a button everything gets done. But there's just lot of human effort that went into this. And we didn't stop just at environmental assets. We went and also got the most authentic clothing from India and worked really hard on scanning them and getting them into engine which means we needed to build our own rig. And this is where a lot of people have this very myopic view of engineering, right? Either you're an engineer that just writes front end code or whatever. Like we think that's so myopic and I've said this for so many years because what we did was we sliced an office chair in half. We turned it around. We motorized it so it can turn. Then we put a wood plate on top of it. Then we put markers so that the scanning software can understand the scene better. Then we put a mannequin on top. Then we set up a perfect light box that only us as a video company. I mean other people could possibly do it but it is it's hard to get that perfect photogometry lighting. You need perfectly even lighting uh so that there's no baked shadows that you can kind of see in the scene or bake lights that you can see in the scene. We had to set up you know an arm. We had to put the camera with uh you know flash and then a polariz on top of it so that it's not able to you know pick up too much of the reflective surfaces. We had to put all this together and automate the entire process as we moved the arm and automatically hit the flash as the cam as the um wheel kept turning. It was I mean I don't even know how to describe it. It's just it's such a good example of Indian jugad. The ability to do things even when we might not have the means or the resources. Very hard to get a rig in India. And honestly, it's not that we couldn't afford the rig. is that it's just impossible to get it down in India and we'd have, you know, hell to face

### [15:00](https://www.youtube.com/watch?v=9Csd_qZyt3c&t=900s) Segment 4 (15:00 - 20:00)

with customs. So, we just decided, let's just make it ourselves. Now, the scans alone aren't enough. You actually have to create the other maps for this object. So, you need to now create the roughness map. You need to make sure you figure out how shiny it is. You need to create the metallic map. Uh, if it's got metal parts, you wanted to actually make it seem like it's there in the real scene in game. Uh, so it's a lot of work, post-process work, even after you got the scans done. And for a lot of the smaller scans simply because we're scanning a huge volume of clothes, we just went with phones. As long as the lighting setup is good enough, it's not that the actual equipment is the blocker. It is the fact that you need to get the setup well and make the entire process happen. Then the entire process of rettopology, which means we need to create these scans and we need to make sure the mesh structure, which is absolute garbage that comes out from scans, works in a game. Which means when you're actually simulating these objects in game, you're simulating a doti in game, they actually feel and look real. And sometimes you have to bake them down into simpler structures, right? So a lot of work that went in making sure these are game ready and then composited on top. And then we also scanned human faces that actually exist in India, people who actually exist in India and added our own flare on top of them. We will recreate the entire city of Chanderi in the game. We're not going to call it Chanderi just because we want we have a story in mind and we're going to retell the entire story of the Ramayan and we didn't want back in the day because somebody or the other will misinterpret it and we just wanted to have the freedom to tell our own story with while being super respectful of the original source material. So we sort of did a loose retelling of the story in modern India and you know there's parts that are there in modern cities and I know there are lots of people that have been talking about you know showing an Indian city in a game for a very long time a high quality city the best method was just to scan it and put it in because scanning got really good in the last few months it's just a herculan effort to pull it off and requires a lot less of sitting on the computer and a lot more of going out and blocking the entire town. We literally had to block the entire town and get permissions. It was I don't recommend it anybody to do it again. But a game is not just assets. If you just stop at assets, cool. You're an asset library company. But a game is more than that. When I grew up playing games, there were specific instances, specific bosses that I remember fighting against that I still remember to today. I carry those memories with me. And I feel we have this responsibility, right, if we are making games to make characters and personalities that people carry for the rest of their lives as a memory. So we wanted to come up with the best story, the best game design. And while I can't spoil too much of the story, we actually put a lot of effort in coming up with a story that everybody will resonate with. A story that where you start with, you know, where most of us are in life and then go through the quest not for strength directly but for wisdom first and then for strength. Kind of like my life. I had to take a pit stop at the wisdom, you know, restaurant to then be able to carry on with my life and gain strength and power, right? I think that should be true for everybody. And the game not only is an excellent combat experience based on Sakiro the combat systems, we've also built a lot of things from scratch that make it uniquely Indian and uniquely something that doesn't exist on the game stores today in any game store because it's just the way we thought about this. The approach that we have is a complete outsider approach. So it feels and looks different. It will feel a lot different from everything else that hopefully has come in the past. And it's not, it just doesn't even end there. It's almost like a never- ending project, right? We had to put in so much time and effort actually doing animations. So, you can take a look at some of the mocap BTS that we have here. Uh, it's a lot of sitting in suits. The animations from the suits aren't perfect. So, you need human animators. You need real people who are going in there and exaggerating the effects a little bit, making sure it looks real. It's a lot of effort. And I know I've said that a million times, but if you knew how much effort this was going in and how much money it would cost, we probably wouldn't do it, right? But it is what it is. Foolishness is important for you to get in. But once you get in, I feel like now we've just made so much more progress than everything that exists. So there's a lot of manual effort that's gone into this. But lastly, in order to make those combat systems I spoke about, in order to make the gameplay feel very responsive, in order to create our own behavior trees uh and not have to rely just on the Unreal Engine ones to make ours goal directed, we had to write an obscene amount of C++. You know, there's this thing that game developers say uh when front-end engineers talk about performance issues, we call them jokes. All of us call them jokes because we just feel like the amount of effort, every strand of hair we need to worry about, right? Every single vertex we need to be like that's too much. We got to worry about it. Number of characters in a scene we got to worry about it. So I really feel like there's a lot of optimization work that we had

### [20:00](https://www.youtube.com/watch?v=9Csd_qZyt3c&t=1200s) Segment 5 (20:00 - 25:00)

to do. And I need to come back to the idea that even after you do all of this, even after you figure out all the programmatic parts, there's still so much manual effort. Like I want to show you a picture of an auto straight from a scan that we had to fix or clean up. It's just so much hellish work. You can look at these fruits that we had to sit and manually clean up. It's just a lot of work, right? And parts of the scans are missing. Like this just I mentioned this like 50 times, but it was a lot of work. And I'm glad that we reached this end of the tunnel where the game is actually 60 70% done. And now from here to where it goes, I think it's a lot easier because we've already processed so many assets. We've already fixed so many assets. Uh there's a huge number of modelers on this. Like I mentioned, there's 40 people on the team. So before I move any further, I want to quickly talk about the team because this is not a indie shop where it's just one or two people doing it. It's a large team working on it. Maybe not the size of most AAA companies, but we do feel that most of those AAA companies are still, you know, using old flows, right? They're not they've not moved to photoggramometry as yet because they haven't figured out it's gotten good in the last 6 8 months. They haven't kept track of this space and they're mid project. So it's hard for them to switch out. So they will become smaller over time too. And I feel like size when you're that big, when you're 100 people, it's a little bit hard. At 40 people, I think we have just the right size. We're still hiring another 20, 25 people. There's still a lot more work left to do. But I quickly want to introduce you to the team. So on the programming side, the person leading it is Rishi. Rishi has extensive experience building a lot of games by himself. He's actually very young, but he's built a lot of games by himself. And he's also built a game engine even before he walked into these halls. And now he's written hundreds of thousands of lines of C++ and it's just been a treat to watch. He's built a lot of systems from scratch and you guys will enjoy playing the game. I hope you guys enjoy playing the game as much as he enjoyed building it. Next we have Davi who is actually from Think Tank which is one of the number one schools for game development and VFX across the world. Uh if you go to Think Tank's website, you can see some of the work she's made and she's a great 3D artist who likes working with her hands, who likes building things from scratch, and you'll see a lot of her assets in the game. Then we have Suge who's who goes by Hansy Sujay and Su has worked on an extensive range of titles from Hogwarts Legacy to Horizon Forbidden West to even go to Sushima. So SU has got seven years of experience doing character modeling. We needed someone senior there, someone who's an artist who has an eye for all of these things and so has just done a fantastic job. Right? The entire pipeline, both the photoggramometry as well as the handmade pipeline is both run by him. Uh then we have Krishna who is the animator from Ant-Man quantum. Krishna has actually worked on Ant-Man quantum mania. So you've seen Ant-Man quantum mania, you've probably seen his work. So we have him actually going in fixing the animations, thinking about how the cinematic should be structured. So it is a pretty good team. Then we have Din whose you know experience and credits who used to work at Dag but his experience and credits include films like Mickey 17 and Aquaman. I think that every India already had the latent demand or the latent need desire to actually go out and make a game. It's not we didn't have a skill issue ever right. I thought I think it's mostly the fact that we just didn't have either the courage or the ability to put things together see the grand vision, the ability to spend so much. Um, and of course without photogometry, this will have to be a much larger team. So a much bigger undertaking. So I just feel that uh I'm so glad these guys decided to come do this with us. We spoke to so many 3D modelers who said, "Oh, no, but this is good, but I still," even though we were paying, we were offering more, they were like, "Oh, but I want to go work at this bigname 3D, you know, a big name outsourcing shop in India, right? " So, it's very hard to get people to take a risk and I understand why, but uh I'm so glad these people took a risk. Then we have Pari who even though Pari doesn't have big titles to his name, you can see some of his work online and he's done an extensive amount of work and Par was the hardest worker because he had to catch up to the rest of the team. So, he's done an excellent job at cleaning up the photoggramometry scans and leading that team and making sure that, you know, the assets that come out in the levels and the set dressing that come out in some of the levels that you saw in the trailer and some of what you'll see on the website, he's put in a lot of effort there. And it's almost like when you're playing catchup, you're like, I want to be as good as the best and he always had the potential. He's always worked by himself for years and now finally he get the opportunity to take it to the world. And then of course the person running AOS games because I do AOS and my job is to keep looking for the next interesting thing that we can potentially be the best at spend 6 8 10 months in this case probably 2 years spearheading the first good product back then it was the first few good videos set the process but the CEO of this company is Rohan who's my brother who is XY combinator and we made games before we've always spoken about making games

### [25:00](https://www.youtube.com/watch?v=9Csd_qZyt3c&t=1500s) Segment 6 (25:00 - 28:00)

we made overpowered card game last year uh which is probably defunct now but before that we've made you know small games that we used to constantly have fun doing we built something called Mumbai make harda many years ago you can find it on it. io IO uh we always wanted to make a game. We knew it was very expensive. We knew that the potential ROI of this is zero. Uh but someone's got to try and so glad he's leading this because I can really count on him. So there are a few more people I want to talk about. One of them is Vishal who used to run this VFX cinematics firm called 20s. He's done a fantastic job of uh sort of making cinematics for clients with VFX. And when we gave him the shot at making cinematics for the game in engine, he said, "This is going to this is a dream and I would love to do this. " So, he's done a fantastic job. So, you've seen some of his work in the trailer. You'll probably, you know, you you'll see some of his work online. And as time goes by, as we leak more things, you'll probably see some more of it. And then finally, we have who's helped us with a lot of the coordination. And then apart from this core team, we have a huge team actually sitting down doing cleanup, doing art, doing modeling who all sit between below the people that I've spoke about so far. And of course there are people who are not directly involved but sort of involved which is we have so many people here at AOS who are actually helping us with things like audio like we find the right person because we've got video and content people here. So audio, audio mixing, video. We have so many people here that are helpful. And uh yeah, I just want to say that at this point we know 100% of how to make a super high quality game. Like we think we can hit the global benchmark and this is iteratively going to keep getting better because 3 months ago if you saw where it was some of the people that saw the demos they thought it was good but now it's even better and it's going to keep getting better as we infuse more of India as we make it more Bollywood style. It'll be very unique. Nobody would seen a game like this. And yes, we're going to try to get it on PC and consoles. We'll figure out in what order and you know, we'll figure out all those things later. Um, but we are also humble enough to know that nobody can predict success. We can try, we can give it our best shot, but nobody can predict success, uh, financial success, but we think we know how to make it. the entire sequence of things. We know how much effort it takes. We know exactly how much it costs at this point. We won't breathe easy till it's done. So ladies and gentlemen, that was Lean. It's called Unleash the Avatar. It'll come fall of next year, 2026. And I'm so honored that we were able to do it. I can't leak too much about the story, but uh but you'll find out details soon enough. The best thing you can do is really share this. I know that making a game is too expensive for most people or it's too hard to put all these things together or you know skills might be a challenge and we're always hiring if you are skilled and you want a shot at this but we do think the ways to supporters are really sharing it far and wide. The hard part is not building things anymore even with a game. Sure it's very difficult and we've moved a lot of things to make it happen but it's also about getting enough of the reach. That's why I had to spend so much time hyping this because the project's risk is so high that if we don't get enough exposure then and this doesn't go to the markets where it matters like the US and Canada and places like that then we probably don't have enough of a shot of being a financial success which is which means I lose all my money. I'm a bootstrap company so it is financially stressful but u I'm hoping you help us with that. Anyway, thanks for tuning in. Bye.

---
*Источник: https://ekstraktznaniy.ru/video/12014*