Most MP asset stuff is trash with bad materials.
Some even go as far has having 10 to 20 material slots on a 10cm^3 mesh.
If you make a game that’s even remotely good, you won’t be using any marketplace asset “as is” anyway.
And any programmer worth a penny will be able to tell you how stupid it is to sacrifice performance for an object you would barely even notice during gameplay.
So @op account for having to re-do or fix all assets you purchase anyway.
And/or contact the asset maker and hire them to work for you.
At the same time, just because you make assets, please realize that maybe you know nothing about making assets for a game:
If I had 1c for every time someone said ‘yea sure, I’ll send you the models’ and subsequently proceeded to delivered sheer trash that wasn’t even UV mapped, let alone light mapped… I could potentially have paid a dedicated person to take care of it twice over by now.
(This is most common with Architects, as they have no clue what rendering outside their niche software means).
In the end, unless you want to become a generalist and learn it all, you have to probably hire a generalist.
Literally - someone who has done and can do everything himself.
Or at least comes close to it - maybe they don’t know networking, but if they know the rest and theory they can probably tell when whoever you hired to do the networking is giving you trash net code, for instance.
Regarding the scope/size.
I just disagree with any limit in principle.
I don’t care how long a project will take. Or the fact that it will be “dated”.
For one,
You cannot possibly cause a high poly / impossible vert count model to ever become dated.
And if you have self scanned 4k sized textures you’ll be hard pressed to ever find those become dated.
For two,
Game play systems are unique, as such its hard for any of them to become “dated”.
We ALL still mimic games like Thief when working with shadow/rogue like games.
Has this now become Dated? Not really.
Can it be made better? Sure. Always.
Same principle can go for another 20 or 30 game mechanics we still use to date.
For three.
Every year your code needs a complete re-write anyway. You learn new stuff along the way. This means you can increasingly do better no matter who you are.
Also, everyone here is correct.
A smaller project scope will make development with less people more plausible.
Deal with it by prototyping in a non prototype fashion.
Where the code and stuff you make is final quality with over-thought optimization, so it will almost never need backtracking and changes.
Just Modular-palooza the s*it out of everything. Animations, interactions, gameplay stuff. Everything.
And do not ever:
- rely on engine features.
- update engine version after starting…
I would go as far as contacting Havok and paying for it. Yep, even as an indy project.
Re finding actual people.
Trust no-one.
Period.
Especially if you don’t know. Have them earn trust by proving their stuff works…
There’s a ton of scammers who claim they know how to code but can’t even add 2+2 out there.
And it’s getting worse because of job cuts in the US…