Ue4 vs Blender Logic

So, I just tried out ue4, and gave it a in depth testing…

it’s so constrictive!

In blender game, I define a input, and relationships, like who the player is, and how he accepts input,

in ue4 it’s like you have to do everything a certain way…

I am now convinced that if blender game had the graphical capabilities similar to ue4 I could build
better games in blender 10x faster than in unreal.

I may have to use ue4 for work soon… but I don’t want to :frowning:

Blueprints are flexible, but in an strange a twisted manor…

1 Like

You will get the hang of unreal.
I love it but I have to have blender to create my models.

ok, I found what I need to get direct access to input events,

Controller - Actor - code container is present (the system I designed!)

it appears that all coders do think alike :slight_smile:

however their abstraction system is a whole window,

my abstraction system was a game object that I could edit the python behind it in seconds, however the same function is there.

one system I don’t get yet, is how to build a variable that is a list that is generated each frame *

after I get that going I will be able to encasulate control myself, and then encapsulate game object interactions…

I am going to wrectifiy Ue4 to make game for a studio it would seem…

You can even make an online game with the blueprints in UE4, you can’t really compare logic bricks to what unreal has.

So does that mean that you wont be helping with
funds to develop the bge anymore.And you will be just making money for yourself.It is your decision.

I will help blender, and bge, and use ue4 for work,

I am being brought on as a advisor,
but I am already wrapping my head around blueprints,

I think SCA nodes or S(set)C(eval)A(get) would be better*

the reason? flexiblity…

UE4 is coded to use systems they integrate, and though efficient, it’s not near as flexible.

a few more upgrades to game logic and I honestly think the bge is a better tool.

Just don’t itterate over large data sets in python and its quite fast.

Aparently not very many operations in game logic can be threaded do to the sequential nature of games, only things like Path finding, and volume checks and things like that can be accelerated, inside the engine under the hood.

Arrays, Trees, etc are very necessary for any core system that operates on a large data set*

(Like making Level of detail, or Lamp hook managers or light managers etc where you need all objects in a radius with ststic elements*)

I haven’t time to elaborate too much here, but, to say that Blueprints aren’t as flexible is simply ludicrous.

SCA Logic Bricks work. This is accepted fact, and we make games with them all the time. Certainly, we all started with them. But the SCA paradigm is severely limited, in terms of scheduling, reusability, complexity, data flow (as you’ve noticed) and many other aspects that BluePrints exceed in every way.

Blueprints can be made to approximate logic bricks. The same cannot happen in reverse.

you can’t pop a python node into blueprints either*

‘Starting out’ in logic bricks’ is much less convoluted than blueprints…

however you top out much earlier in SCA however you can extend using python,

if I were able to store a property using a sensor, and use it using a actuator, this would raise the level you can go with logic before topping out*

adding more logic bricks would expand this area as well, like

Ray(set property)-------------------------------------and-----------
if property is game object and valid-------------and-----------Steer to target

and when I thought blueprints was less flexible, I was going through the input handles they use in the FPS demo, now I know you have access to input->keyboard-> key events

they are not listed inside events and it was scary**

You might want to check out Skookumscript too. It was released on the marketplace recently and is a free plugin. You may like it if you enjoy python scripting (although it doesn’t really look like python).


http://skookumscript.com/about/

I want to setup my own input control and it looks possible,

however I need to be able to store a value on a list if that value is not on a list each frame, and then evalute that list later and act upon it,

I will get back at it tonight after my midget crashes.

reiterating = Inputs should be under events in a sub directory* it scared me when I did not find them*

their pawn system is good, but it’s not what I need specifically*

UE4 is amazing, but the ui intially is quite intimidating*

1 Like

Now to replicate the property sensor…

Looks like I have to bind a property to each instance of a class, and then make a custom event,

if property !=0---------trigger

edit: if I can get this working 1 to 1 (game logic to BluePrint and SkoomaScript to Python… I could make a exporter…)