Feedback Future of unity

seriously unity should start taking big steps and that includes only releasing 2 versions a year (alpha and full version) so we don't get buggy features and spend hours if not days trying to find a good workaround for an issue that should've been fixed before publishing the version, it should remove all the unnecessary old stuff like old ui, old input, old components, methods and classes, also it must make major updates like adding the 64bit transform precision support (which everybody was asking for since 2011) cuz rn you can't have an object at more than 10k units without jittering and getting the light totally messed up, even godot supports 64bit transform precision...and i think it'll be cool to only have to download the core engine and then just add the packages you need through the package manager

also please finish those preview packages cuz c'mon they've been in preview for 2 years

At the end i wanna say to unity technology developers "Please stop giving people bad workarounds instead of fixing the bug or adding the feature that about 80% of the developers will need"

 

Christ.

1. There's only going to be two point releases per year going forth. Going from alpha to release is a terrible idea and not something that will magically reduce bugs.
2. Deprecating old features en masse will not make the engine more stable, it'll break thousands of projects and push people towards new features that might not cover the use cases they need.
3. 64 bit precision causes a lot of fundamental problems, especially if you need to make any use of GPU support for calculations because you can not rely on those calculations on all hardware. Solutions to this problem exist already and are not terribly difficult to implement.

 

Out of all of that, I am just sitting here wondering what a UWRP is. (Universal Waiting Render Pipeline)???? lol.

 

Game development supposed to be full of choices. How people are developing games if they can't make a couple of simple choices (risking new stuff or go with the good ol' comfy)? I think I will never know.

But admitting the need for the "one button game development", even if it's only in a picture is cute.

 

When you tell the wannabe MMO creators that "make a MMO" button existed in 2016, they're going to have a heart attack.

 

On a more serious note:

This software is huge, and there will always be bugs.

This will alienate 70 to 80% of user base and cause massive profit losses, if not kill company outright.

And this is actually difficult from technical standpoint.
Not a whole lot of people has been asking for this feature, as it is only has very limited utility (you want to make a planet. You don't have the budget for that anyway.).
Physics system does not support it, PhysX does not support it, GPUs don't support double precision out of the box. Unreal engine does not have this feature (think about it for a moment). So it is a big time investment, and performance loss across the board.

And for what?

You can achieve comparable results with origin shifting.

 

It's not something I've tried myself yet, but HDRP supports Camera Relative Rendering, which helps with the world origin issue.

 

I've been using Unity since 2007 (v2 if I remember correctly). Back then, unity was an incredible piece of software, lightweight, very well designed, and Mac only! It wasn't playing in the AAA category but despite not having all the eye candy fancy features, it really was a solid tool to actually release games.

With each new version Unity was getting stronger and gained a lot of traction. But around v5 things started to degenerate seriously. They began adding new features after new features at an incredible pace without ever finishing them.

Now, Unity is a TOTAL mess. Things are going in all directions, URP, LWRP, old UI, new UI, old Text, new Text, packages, old special effects, new special effects, enlighten, assets, alpha versions, beta versions of multiple versions at the same time, etc etc. It never ends, it's bloated as f***.
In this current state, I wouldn't take the risk to develop another game with Unity. Last time v5 barely made it without breaking too much things. But now Unity is not on solid grounds anymore.

So for its future, I wish Unity goes back to its fundamentals, stay true to its DNA.
Please, make Unity a simple, elegant, reliable and powerful software as it used to be.
Integrate basic tools that run great, that make the job done and pay the bills in the end.
Why running after all the bells and whistles of UE5 that anyways no one REALLY use in the end?

 

This happens to a lot of software, actually.

I remember Visual Studio Express ... 2005 and 2008. It was relatively lightweight and robust, and you could disable most components.

Now... Visaul Studio 2019 is a behemoth. You can't remove things you don't want, and it hogs resources to ridiculous degree.

 

A lot of crying on the forums lately over software most of you are getting for free.

 

Imagine having paid for it and you're not even allowed to cry

 

My experience of URP and 2019.3 has been pretty solid.

i just released an online multiplayer async game on android and ios in 4 months... by myself. I couldn’t have done that in unity 3,4,5.

maybe that’s my experience coming through but I honestly think it’s easier than it use to be.

sure there are bugs and quirks but there always was. Trust me, i sunk days into tracking down performance issues before the profiler was a thing, remember that? No solid profiler! Things like this are taken for granted now.


Remember in unity 3/4 how there was no way to order the execution of scripts? And how the order would vary from platform to platform? These things are so easy to deal with now.

plugins use to be a mess.

things like ads and IAP were really hard, now it’s possible without code.

so many improvements that fundamentally improve your game’s business performance

even setting icons on Ios and android was a pain in the behind, but that’s now totally friction free as well.

Custom inspectors, property drawers, custom editor menus, so much useful productivity increases and these things only continue to get better

 

Geez, I must be using a different engine. I just moved a project from 2018.4 to 2019.3 and had no significant issues. I even enjoy no longer being logged out of my account in the editor every couple of days for no reason at all.

I miss the good of old days. Back in 2009, you backed up. A lot. It was always a gamble whether you would get a solid hour of dev in before the editor crashed and corrupted your entire install. Kids these days.....

 

There's more issues to the engine than just stability and API compatibility between the most recent major version and the last LTS.

 

Indeed there are. But the sentiment being expressed above that things are a mess is hyperbolic. *Especially in the context of historical experience. Five years ago I was at the front of the line of torch and pitchfork wielders.

 

Yeah, URP is really nice for some types of games right now, especially 2D games. I was messing around with a 2D top down pixel art prototype recently, and I was extremely impressed by how well URP delivered. Fast, reliable, etc. The 2D lighting effects and 2D post processing effects work great.

 

FIX UP COLLABORATE!!!... No improvements to Collab for almost 3 years!!... it has the potential to MESS UP YOUR PROJECT.
YOU STILL CAN"T MERGE SCENES! LOOSES FILES!!!! TOO SLOW ect ect. JUST DOESN"T WORK AT ALL...

 

Sorry, but going "historical context" only works if history remains a fixed point in time, but there are multiple engines out there that are really starting to eat Unity's lunch. GameMaker Studio is now an incredibly robust option for working in 2D, Unreal is seeing a lot of usability improvements even if it still has a ways to go, and even Godot has made strides towards being Unity without the cruft.

Unity remains a good value proposition for my particular projects, but it's not hard to see why a lot of people are seeing otherwise, especially with how fragmented Unity has become thanks to its big focus on external tools acquisition and incredibly fragmented editor tech and usability.

 

Repeat after me. Subversion is your friend. Subversion will fix all your woes.

 

Please, find a hammer. Make sure you still have your caps lock on. Then hit it with the hammer with everything you can muster up. Thank you in the name of the entire forum.

 

Yeah right..Another messy solution. eg. Would this allow multiple people to work on the same Terrain (Different Sections) at the same time? No. Merge Scenes? No

 

CVS anyone?

 

Yes. Since it's different sections you just need a way to separate them. I haven't tried it but I wonder if the current prefab system wouldn't work for that.

A scene is basically just a fancy text file. I wouldn't be surprised if you could merge scenes but I wouldn't recommend it.

That said this was mostly intended to be a joke. Git is a far better recommendation for someone unfamiliar with version control software. Svn is one of those things some of us keep using because we learned it and it's good enough to get the job done.

 

Pointless ranting, wrong forum. Closed.