I had previously developed a AAA quality game in 7 days(link), during its development I realised the 

importance of optimization and straight away from thereafter I have been studying about optimizing 

games especially in the Unity game engine.

I tried searching youtube about how to optimize games and tried many things on my game that doesn't 

resulted in something very good but I had idea as what to do and what not while optimizing a game.

That project finally broke and my machine wasn't able to bake lights for it or even run it at 10 fps.

I started with the Boat Attack to initially study URP and it's optimization and found out that even it is

not optimised for mobile, it ran at 13-15fps on my android which has SD 730.

From then on I started optimizing it and realised that the optimization is really shady field even to

the study, there are very limited videos but lot of documentation to read and the numerous hours of 

reading that only helped me.

Initially, I realised that Vulkan architecture was the problem and disabling it boosted the fps to 28-30.

I raised an issue on it's GitHub repo also: link

The profiler screenshot is given below:

I tried hundreds of things which I can't even remember, almost triggered every setting available inside

Unity from rendering URP settings to Build settings and finally I was able to achieve 50-52 fps on SD 730.

The final build optimised apk link is: here

Do watch how I made a game in 7 days video here(link).

I tried my hands on with the Unity's Shadergraph in Unity 2019.3 in Universal Render Pipeline.
I started off by creating a simple flag wave shader using Unity's learn live, then went on with exploring YouTube and creating more shaders and getting used to the shadergraph experience.

The major difficult thing was the URP, although it has shadergraph support but it lack features when compared to HDRP and that proglem was with almost every youtube video, although Brackeys were really good at this.

Most difficult shader was the ocean water one, that has to be created in URP but again had multiple problems like the depth node was most difficult one and now also I fear to use it, even I don't know but somehow after creating water for 2 days(that too following a tutorial) I had water, ha ha.


Complete Project Github Link: https://github.com/sanchitgng/ShaderGraphURP

Space Apps Internationals but those represented Apollo 

Mission sites and here’s the catch that I got, why not 

make a moon explorable terrain that involved India’s 

Chandrayaan mission, it was trendy at that time as it 

was about to land on September 7, this hackathon was 

on 16-17 August as far as I remember. So we just did it, 

Rahul did the research part and I just quickly implemented 

the Lunar South Pole of Moon creating landing site of 

Vikram Lander with our lovely Pragyan rover created by 

Praddyum(he was in another team though ;)), he is our 

3D modeller. In 2 hours the prototype was complete and 

in approx 2 hours more, the results were announced and 

we had Won! Just like that, I wasn’t expecting as this was 

my first Hackathon but luck was on our side though.

We qualified for Nationals of Space Apps to be held at 

Delhi and later we came to know the date, it was 18-21 

October 2019 and so we started again, the problem 

statements were out and there was no Moon problem 

statement, so we had to make again from scratch and 

this time I did some real sweating work and Rahul, 

some real research as we had to plan a journey to 

Kuiper belt that should be scientifically accurate in 2045, 

Yes! We had taken some assumptions -3 assumptions to 

be specific. Rahul planned out the journey and I did the 

implementation part, it was during the Durga Puja and so 

I had the time and I seriously used it. During the 36 hour 

hackathon, we had made the full journey that would 

start from Earth and take you to International Space Station 

which had some real no gravity effect and was exact replica 

of the real one, from ISS you are headed towards Kuiper belt 

objects namely Pluto, Makemake and Haumea, if you are 

wondering what are those weird names, they are just names 

of planets with some special behaviour, I mean absurd behaviour.

We could just tell one thing that the quality 

we had made, the graphics was at par for the mobile platform 

but still we LOST! as it lacked innovation and yes it lacked!

The hackathon was over but SpaceVR was still on and so this time 

we took it for a B-Plan in KIIT-Fest in December and lost again but 

as it is said that there is always gaining and so we now had a real 

business plan for our SpaceVR that could be presented.

We expanded the project more and this time we designed complex 

scientific phenomenons in VR, like gravitational waves, spacetime 

wells, accretion disc, gravitational phenomenon, planetary movements. 

We were making this because our app was moving forward as per our

 business plan that was to make educational content in VR, generally, 

such type of education is considered for elementary schools only, 

therefore, we decided to build complex scientific phenomenons like

above so that our app focus would be on serious potential educational 

content for senior schools that has not been created before.

We took our project to IIT Bhubaneswar this time, it was AR/VR 

conclave organised by IIT and professors from all IIT’s IIIT’s and 

IISC Banglore had come who were working in this field, it was a 

hackathon and we were judged five times. This time we Won! 

Although it was 3rd position but it was IIT, the tag that everyone wants!!!

Winning losing is always a part but what is most important is the

learning and the journey, and every time this project Won was 

when I was focussing on my learning and least expecting it, 

the process rather than the result, during our Nationals of 

space apps I was too much focused on making something 

that could make us Win this time too! but as you know I Lost! 

But a lesson learned for the lifetime that I already read somewhere, 

in Shrimad Bhagwad Geeta! -focus on karma rather than the fruit.

Don't forget to subscribe to my blogpost newsletter here

The project links are: 

Project ppt 1

Project ppt 2

Business plan: ppt Link

Watch my video of making a game in 10 mins here

The official transcript published by KIIT Today:

Here is my 2019 compilation video of ever game scene I developed. Hope you all like it, if you have questions or something to say about this video or of any of my projects, you can do so by comments!

Is Being a Good Game Player necessary for Game Development?

1. Choose a game engine: As you are a beginner in Game Development, it is best to 
   start with a known Game Engine. There are many available such as Godot which is free 
   and open-source, Unreal- which is used at the industry level to create stunning games 
   and Unity game engine which is used by almost everyone from indie developers to 
   industry level players in the market, from AR/VR to real-world mobile apps, all made 
   with Unity. My personal preference is Unity itself as it provides all tools to get you started 
   as an indie game developer and you can also make real world AR/VR mobile apps and if 
   you are too good at Unity then you can use it to make almost any kind of app that you can 
   imagine!
2. Create a basic game: There are now several official Unity YouTube videos which can help
   you get started with creating your very first basic game, don’t try to create anything very big
   at first, in fact, create something very small, it sounds weird but you need to hold yourself 
   to give a good start and once you are a pro at Unity or Unreal you can open the fences and
   imagine anything and try to create that one.

My very first ‘playable’ game was Project Boost, a 2.5D rocket thrusting game with two levels.

3. Move on to 3D: When you have created your first basic game which should probably 
   be 2D or 2.5D for it to be called basic then move on to learning 3D concepts on game 
   engine. My second game was a basic 3D ship flying game which I learned through a 
   course on Udemy by Ben Tristem and Rick Davidson, believe me, they are the best 
   teachers.
    

4. Learn 3D Graphics and Associated Maths: It is not possible to make a video game 
   without graphics and so you need to learn about them now. 
   The maths commonly required for 3D Graphics is Linear Algebra which mainly deals 
   with linear equations, matrices, vectors, etc. 
   If you understand these concepts, you can easily manipulate the 3D world of your game 
   to achieve miraculous effects!!!
5. Publish a game: It might not seem how important this is but you need to publish your 
   game asap, anywhere publish it, it is not to be on PlayStore or Apple Store, publish it 
   because it gives you a push onto your learning and game-dev journey, don’t wait for your
   big awesome game debut, that’s not how it works, publish anything small but you should 
   have a good feeling about your game. My first game published on PlayStore was 
   Mad-Ball Breakout and that’s the only 2D game I have ever tried to build because 
   seriously I don’t like 2D as I haven’t been a 2D game lover myself.
    
6. Learn random concepts: Once you have the idea of the game engine after at least 
   six months or so then you can just learn random engine specific concepts like 
   Dynamic Scrolling, Optimization, Post Processing Stack, etc.

Here is an image of before and after applying Unity’s Post Processing Stack features,
 

Before Post-Processing,(below)
  
how much detailed and good the scene is now, details such as Lens Blurr, detailed fog, detailed and 
darker shadows, HDR, etc.

What are some Online Resources to Learn Game Development?

There are several resources available such as books, courses, and official videos, the major problem
with almost all of the books and most of the videos is that they are outdated, the process of making 
and publishing a book takes as much time that the information in it gets outdated, if you are following
a video try to make sure that it has exactly the same version as yours otherwise you will have a hard 
time figuring things out and will eventually leave it.

Udemy Courses are really good but just see the ratings and reviews of students before actually 
opting for the course, but by far I think Udemy is the best bet for now and then comes YouTube, 
you can always go on and contact me.

Sanchit Sharma

www.sanchitgng.in