How to start a project??

Jordanbeeston

Hey Guys,

Im starting a new project and I really want to start it off properly this time and advice i have heard is to plan out the level and assets needed, im kind of half way there but struggling on what to put down so that i can get to work... or should i just stay with what i have and trojan through the assets one by one?

here is my level blocked out in ue4:





i also have a screenshot of my excel asset list, is this the extent of planning i need or would people do it differently?



its all a bit messy to me but to neaten it up will take time, time maybe i should be using for actually creating the assets?

Another thing i needed to look at was which bits should be modular and which bits are hero assets, i think perhaps they could all use some kind of tileable texture though but im not sure at this stage

sprunghunt

You shouldn't worry about process so much. You could start by making one asset and then going from there. But it doesn't matter how you start. What you want to avoid is over-thinking it. Especially when you don't have a lot of experience.

https://en.wikipedia.org/wiki/Analysis_paralysis

BUT - I notice you've got the buildings listed as one asset. You probably shouldn't do this. Make the buildings in pieces (windows, quoins, gutters, bricks) and then assemble them from there. 

lotet

First rule of solo/indie video game/project development. half way through you’ll realize its sucks and you have to redo everything anyway.
basically whats gonna happen is your gonna learn a lot of new things that you want to apply to the first assets you made, and now you have the power of hindsight thats gonna help you re-plan everything with a munch clearer sense of direction. 

the sooner you embrace this the better, it will help you relax since you know you’ll have to redo it later on anyway =P

sacboi

Got nothing more to add except chill out. Re-editing and optimising as your project progresses is all part-'n'-parcel of game development, something I've personally only come too realise myself.

So go with the flow bro

r13

make one thing... dont worry about anything else cause when you have it working, you save it and go to the next item.   it seems simple, but it's how every game project is done. small steps. each step isnt very hard, the trick it to not get tried of many small steps before you can realize your goal. learning to code, making art, making games... its all done the same way... small steps that you build on to get to where you want to be.   

sounds like hallmark bullshit, but i assure you it's not. take it at face value, no deeper meaning, but its also not to be ignored. 

good luck and stay on track

Jordanbeeston

Hey Guys!

thanks for your comments, i know in the back of my head that your right i just need to put thought into action now!

aha i looked up the analysis paralysis sprunghunt and its definately what i do and end up procrastinating for pretty much anything in life! probably wont read into it too much or i might end up over-analyzing my over-analysis! haha

so i shall just start on one asset, guess it doesnt matter which and keep grinding! i might make a progress blog too to force myself to show progress at the end of each day

Thanks again for your help  

SaraSchwartz

What I've learned from trying to do environments is to start big like youre doing.  Get the feel down without actually getting carried away in one small useless prop.  Once you commit to some small piece you're going to hyper focus on it and more than likely you're going to make it too detailed.  

Do what youre doing but now make sure the lighting is set up the way you want.  Don't leave that until the end.

After the lighting make temporary props.  Like block out props, not cubes like your image, more like "almost there" but not unwrapped or anything.  Get the shape down.

After a while you'll poke around with the map and move objects, cut out a few, and then decide based on the lighting and arrangement which props need more attention and detail, etc.

Since some people say do one object first, maybe that's their (and your) way of doing it, but for me it makes no sense to make one small object first.  I don't have a ton of experience professionally but the company I worked for did as I described and I never realized how perfect that method is. 

You can have the shittiest textures and semi shit models, but with the right lighting/mood, arrangement, theme, it will still look good for half the amount of time as doing every small prop 10000%.