I must be one of the rare people using the engine who actually wished it was more bare bones. Now many feats are hard to skip and are hardcoded in a way that you have to reinvent the wheel to make things work again your way.
I think there is a lot of clout to the OP’s post. Tutorials are lacking and one example, in the documentation on the website they show example blueprints, but they don’t show the whole blueprint and just show small combinations of nodes. Its kinda like we are stuck between pandering to geeks and pandering to ppl who need it simple.
What I wish is that they had a hotkey, hover over any button or function and have a description or web link. Even a built in wiki would be perfect. Basically some easier way to get information about functions ect, and on whatever detail level the dev wants. Another example I want to know more about the nodes in blueprint and the official tutorials just barely beat around the bush and the documentation’s description of the nodes is not verbose and at the same time too wordy to understand. We need a wiki.
Hey sorry I hadn’t gotten a response into this thread yet. It’s a really interesting discussion. Of course we are going to try to provide as much useful content as possible, but sometimes it’s not all out the door yet. I see a few requests in this thread that we’re actively working on (VR replication out-of-the-box), a couple of new ones that I’m wrapping up into a report and a couple things that are just unavoidable game development hurdles. I also sent the marketplace feedback to that team to assess.
Tutorials are definitely high on the priority list and we have a lot of new initiatives there to provide more content to the community. For example, you may have seen how we’ve worked with Mathew Wadstien and will be working with more tutorial creators in an official capacity.
There is not much we can do about replication requiring you to watch a lot of tutorials, as the reality there is replication takes quite a bit of work on the developer’s side no matter what engine you choose. We do have plans for better steam integration and exposing more replication functions to BP, but even if 100% of that was already done, you’d still probably need to watch a bunch of tutorials.
There is a lot to address over this whole thread, so I’ll break down the various asks as best I can and get them to the appropriate people.
Thanks a ton for the feedback!
Key features built into the core engine humm I don’t know. Would love to see more “add to project” templates and systems based plugin’s
I cant say I disagree about tutorials, Im at a point where I only use parts of the tutorial I actually require and as much as Id love to have the time to follow along a longer tutorial series Id probably end up using that time working on my own project. Its a shame so many newcomers seem to be avoiding Epics great learning materials as it really is great and thats not even taking into considering the plethora of third party tutorials scattered all over the internet.
I can see your line of thinking here, the Character movement is a good example how the Capsule Collision is hardcoded and requires work arounds, obviously you can use Pawn but those hardcoded features should be avoidable and implemented in the templates/plugins not at engine level.
My concern is the engine working consistently with the features we have and fleshing out of said features instead of implementation of entirely new pipelines or APIs, I know Epic shouldnt have to write our games for us but it is a Game Engine and I feel as though sometimes Epic loses focus of the Game bit in that equation focusing on features for Film and VR which may not always have a large impact for games as they are now (vs how they may possibly be in the future). A majority of the industry still revolves around platforms where VR has little to no impact and there are physics features which could impact games on a whole instead of being specifically geared for Mobile/PC/VR/etc, its those game framework features Im more interested in from a Game Engine because there are so many renderers or other bare bones engines out there I could choose.
I mean I would like to see a future where game design is actually a topic of discussion too not just implementation, why one person might choose a method of doing something one way vs another, its one thing to be able to build a game its entirely another to design a compelling experience. By giving people more tools then system design becomes less of a hurdle and the game design itself becomes the point of focus, its an issue I see in alot of UE4 projects, this could possibly be a good focus for tutorials that would benefit me. One example I asked recently was what do I do about performance of alot (100s) of projectiles being spawned on screen at once? Obviously I have my own ideas but the engine isnt entirely general purpose and has hurdles I have to overcome because of its design, its not something I would face in every engine so its not entirely game specific!
Thanks for replying - for some reason my notifications on this thread stopped coming in.
For me it’s just a little confusing why certain functionality that I would consider “Core” features for a game aren’t included from the get go. Maybe it’s due to my experience as a mod designer but you’d think that a multiplayer server browser with basic Steam integration would be something that most could benefit from. The same could be said about a basic menu and settings framework. Like a template that we could build off of and customize to our own needs. I’d imagine that every game needs the ability to allow players to change monitor resolutions, basic graphics settings and remap key bindings. Those are the features that are making the engine feel bare bones and incomplete in my mind.
Personally I think you guys are doing a fantastic job on the tutorials front. Maybe we just need a quick link to the official youtube channel from the Epic Launcher so more people can discover them?
Yes this is actually what I meant. I chose the wrong wording when made the post. An add to project template for Multiplayer and a Menu system are the two biggest needs as far as I can see. I game isn’t even functional without a menu system so there should be one we can add to a project and build off of.
I agree that the focus on tutorials shouldn’t necessarily be a priority. The community handles this fairly well and if you need help on something, searching “UE4 whatever you need” on youtube has you covered for pretty much everything. There’s a long learning curve for game design and it’s expected. If the odd Epic staff member like Parish and want to make some official content then that’s great but we don’t really need much more than that.
I only skimmed through some responses in this thread but what OP is asking for is essentially a “make MMO” button.
This is the telltale giveaway. You expect to be able to set up a multiplayer system in less than 30 minutes (including educating yourself about it!)?
I say it time and time again - game development has become too accessible.
Also, someone mentioned a key rebind menu as an example of a “core” functionality that should be pre-built. That’s pure BS. A vast amount of UE4 games are for consoles, meaning that sort of functionality would just be unneeded bloat in the engine.
I agree. Complaining that you have to spend 30 minutes to setup basic multiplayer just looks me to, like complaning that sun is to hot on desert.
Nroamlly it takes days or weeks to get multiplayer working.
I’d rather get things that I can really benefit from. Like proper support for Multithreading on UObjects and move from the GameThread hell (it’s 2017 and most of engine and editor still runs on single thread, because when you try to run UObject on something else than GT, it will probabaly explode). And we are getting more cores, not higher performance per core.
In the realm of unlikely I’d also like to get in engine support for double precision on coordinates.
And to bare bones. I just look at actor class and think why it doesn’t make me coffee. I wish we had class above actor, which can only be (optionally) placed in level and support componenets, but doesn’t implement any functionality beyond that. For that matter Networking cloud be move into component. If you want object to be replicated add NetworkComponent or something, instead of relaying on actors to replicate data (and subobjets).
To support my point. Bare bone actor takes 1.5kb of memory. Maybe it doesn’t seems much but it stack up.
Hardly. Firstly my issue is a lack of a proper framework for a Menu that allows players to set graphics options and change key bindings. This is pretty standard fare for every game and we shouldn’t have to pay $20 for that functionality…
And oh yeah sure, key binding code would totally bloat the heck out of every console game… it’d be what? A few megs? Serious bloat!
Secondly I was also suggesting that a basic framework for Multiplayer would also be useful. Not server support, just the ability for players to connect to an online game through a traditional listen server or LAN. In fact they already built all of this for the ShooterGame so why not strip it out and make it into a free content pack that we can integrate into our own projects?
I’m not even making a multiplayer game myself. And I’ve already said that instead of “Core engine functionality” I should have said downloadable templates but since you didn’t read anything you wouldn’t know that. I didn’t even make this thread for my own benefit as I already purchased a menu system and I’m making single player games. I was simply wanting to provide some feedback on how Epic could make the engine more accessible. Bravo for you if you already know how to code everything yourself.
For the vast majority of games it really isn’t.
I was simply wanting to provide some feedback on how Epic could make the engine more accessible. Bravo for you if you already know how to code everything yourself.
[/QUOTE]
As I already argued, the engine is currently too accessible. The vast majority of AnswerHub and support chats are in a constant loop of reiterating how to detect when a player picks up a power-up. It’s essentially going toward the “state of Unity”, where the community as a whole is a perpetual beginner, dwindling down high-level development discussion to a few select threads and private communities. I’m all for making development easier, but a) there needs to be some minimum entry barrier and b) UE4 and Unity aren’t making development easier, they’re making development appear easier to non-developers.
For example, making a turret that aims at the player and has an alarm light flashing on top of it when tracking is pretty trivial with blueprints, but for people who know what they’re doing it was already trivial.
Bottom line is - not is made to be a developer. Or, to use a more leet term - git gud or go home.
You’ve got a very narrow way of thinking. The more people we get on board, the better the potential for. I’m sure you were a beginner at some point but you seem to have forgotten what it’s like to be at that skill level. If Epic didn’t feel it was a priority then they wouldn’t be spending their time on it. And if you’re so good, then you wouldn’t be waiting on Epic to do anything, you’d be coding it yourself because you’ve got engine source code. Bottom line is, you Sir are a Jerk. Coming over to a thread without properly reading everything, misquoting people and then trying to belittle’s opinions isn’t helping anything. If you’ve got nothing to contribute then don’t bother wasting your time that could be spent on improving the engine with your vast knowledge of game design.
I’ve been using Unity for several years and I’ve seen what their “EVERYONE can make games RIGHT NOW” approach to building their community has resulted in. I dread to think that one day this ends up looking like that. I am not belittling’s opinion here, just yours, because I think little of it. You argue that the engine needs to be more accessible, I think that’s inane and that people need to understand that this isn’t MS Paint, but an extremely advanced and difficult tool. Secondly, people need to understand that making games is even more difficult, requiring expertise and precision. It’s not something that one slaps together in an afternoon. This goes ESPECIALLY for the very example I hold against you - “why would I need to spend 30 minutes to create basic multiplayer”. This tells me that you don’t have any clue of the intricacies of networking, replication and multiplayer in general. If you did, you wouldn’t think it a lot to spend months on it, let alone 30 minutes. This is grounds for why I simply do not hold your opinion in high regard.
If you want to label me as a jerk for desiring a minimum expected knowledge requirement, then go right ahead. But you need to understand that “basic multiplayer” (or other extremely advanced “basic” things) aren’t something that anyone can do right from the get-go, and striving to simplify that process further for those people, as opposed to requiring them to learn the existing methodologies, is both futile and stupid.
By the way, I do make the things that I need that are missing from the engine. I also contribute them back as pull requests. I even have a shiny badge and everything!
Lastly, to address the “without reading properly” accusation - you are correct, but just glancing over the “30 minutes for basic multiplayer” line gave me all the information I need. Sorry but not sorry.
A big part of this problem stems from the near entire death of games releasing their own editing tools. Back a decade and some ago we had a multitude of games that had excellent mod/map support. Halo, Unreal tournament, basically all of Valve’s games, to name a few. Fast forward to now, and there’s what? Skyrim?
You’re always going to have people who are just starting, and at this point there’s not a heck of a lot of games they can start out with to get some idea what they’re doing first, so they just dive directly into some major engine.
I just went to fetch the quote and realized I did misread it. He says “go through twenty three 10 - 30 minute tutorials”. So yeah, apologies for getting that quote wrong.
That being said, that’s still ~12 hours of tutorials, which is the bare minimum one would need to start setting up multiplayer, so complaining that those tutorials are required for that is still inane, just less so than what I originally thought.
My other points also still stand.
(The badge stuff was tongue-in-cheek, I hope that went over properly in text)
I wasn’t calling you a jerk because of the fact that you were saying the engine should require a minimum amount knowledge, I was calling you a jerk for the way you were handling yourself in the discussion. It’s hardly beneficial to post simply to tell people that they’re stupid or misguided - especially when you’re basing that opinion based on a misunderstanding to begin with. That said, I accept your apology and appreciated it. I’ve been in the game design field since 1999 so I know what it takes to make a game. That doesn’t mean that Epic - a company that employs hundreds of people with their focus being game engine development - can’t help us speed up the process by making certain functionality more easily accessible. The fact that making a multiplayer system is hard is exactly why I’m asking Epic to build us a framework that we can choose to integrate into our own projects. It’s not like I’m asking them to make a very specific feature set that would only benefit my game. I’m asking for a framework that would literally benefit at the very least half of their “market” if not 70%. Most people would love to be able to have an easy multiplayer framework to work with that could save them a lot of time and money if they didn’t have to develop it themselves. Most people would love to have a Menu system with graphics customization and key mapping that they could customize into their own unique system. The whole point of licensing an engine is that you SAVE time. And we ARE licensing UE4 as we have to pay 5% royalties from all sales. Sure, it’s a fantastic price for what we get, but that doesn’t mean we can’t point out what we feel are flaws. If you don’t agree with my idea of what should be considered a flaw, no problem and you’re welcome to say so. But the way you choose to say it is only going to reflect on you and what you were saying was coming across as really jerky.
Moving on from that argument though, my point wasn’t that it’ll take 12 hours to implement a multiplayer system - which by the way would only be true if implementing the functionality ran real time to the duration of the tutorial which there’s no way it would if you’re learning. My points were:
A: It’s quite simple for someone who works for Epic to implement a system like this because they’re not learning
B: If they’ve already built it out for a tutorial series then why not make it available as a plugin? And
C: They’ve already coded out a full multiplayer frame work in C++ with the Shooter Game example including bot support, but this also isn’t available for to benefit from. My worry is that the reason that they’re not releasing this to the community for free is that they don’t want to “step on the toes” of developers on the marketplace who are already selling this functionality.
Again, I did mention that I come from a modding background so maybe I’m spoiled, but I’m used to having an existing framework to work with and then expanding on it and since Epic has a long history of mod support and UDK was a lot more feature complete even in early versions, it’s disheartening to see the current state of UE4. I understand the reasoning for Epic making the decisions that they have, but it would be nice if we could get some better frameworks to build off of. It’s not something that would take a large percentage of their time - considering they’ve got Paragon, UT4 and Fortnight to pull from, all multiplayer games. It would only benefit those of us who could use the extra support and it wouldn’t harm those who have no interest in those frameworks, they could simply opt out.
If you want to know what a bare bones engine is, take a look:
Almost all game systems there you’re going to make them yourself in order to develop your game.
Also, get ready to write yourself many many graphic shader programs… It’s almost an Unity 3 reborn lol
Don’t even speak of “gameplay framework” or networking and etc.
(Although small engines are great pick for certain types of projects, Unreal has tons of code that many indies will never use)
It doesn’t look all that bare bones to me. But regardless, isn’t it wonderful how many options there are for developers? If UE4 is too “feature complete” for some then I’m sure they can use something else.
I don’t know why people think that having the extra code is somehow detrimental. I mean, unless you’re making SNES cartridge games, I don’t think it’s an issue.
Not sure about the discussions in this thread, but in terms of graphical features and some necessary tools, it really is bare bones.