#.
59 messages · Page 1 of 1 (latest)
hehe noice
@obtuse orchid if you can find a nice screenshot of your shtuff, put it into the "social preview" of that repo, then it will embed that picture rather than the boring githug default
github has a social preview feature?
example 
Both, you're basically describing writing a full blown general purpose engine like Unity from scratch, and as if that weren't hard enough, not using the STL either
Sounds like a decades worth of work
smh just use EASTL
NIH bingo lets goo
I don't see VSM on the list tho
you will be happy once you implement VSM
good shadows make everyone happy
It's just that most people don't manage to make anything even without NIHing literally every component so forgive our pessimism
If you prove us wrong and actually build this thing with everything you're describing and it's actually a usable tool you will be a GP legend forever
I would advise not to build an editor
because that's time consuming and most of the time kinda sorta useless
it is also advisable to try to think of the game you want to make
I'm confused because you sound experienced enough to know how to do these things yet not experienced enough to know that this project as described is like a lifetimes worth of work
But maybe I'm misinterpreting the premise
Yeah but how are you going to spec your engine API if you don't know what it's meant to be used for
Unless it's just a model viewer and nothing more
But your "basic foundation ideas" listed above basically constitutes one of those engines
imo it's still worth to try and design the thing around something, even hypothetical
if you don't have something to look towards it'll be much harder to write a good™️ product
It's even more important to design around an application if it's not meant to be an all-purpose COTS engine tbh
Ok fair enough
we'll watch your career with great interest
Wait how will you have ECS and scripting stuff but not have skeletal animation
welcome to the NIH club
Skeletal animation is like a very basic requirement for any 3D game with animated motion
ECS and parallelization and NIH STL stuff is all way way low priority by comparison
what demon is saying here is that skeletal animation is extremely important and, by comparison, easier than scripting and rewriting the STL
What sort of bottleneck are you hoping to avoid in std::string_view to implement your own
I don't really see the utility
String view doesn't even allocate
You mean like different size?
Why is that an issue
Order?
I see
Yeah I guess serializing STL types is pretty cursed so that does make sense
+1 for using c# for scripting btw 🙂
pascal case, makes the code not look too horrible either
noice
if FunctionHandler was named Delegate xD
all good 🙂
it kind of looks right to me tbh
can you provide a clip, or a picture with ssao off and on?
showing the final output, not just ssao itself
hehe yeah that looks better