PDA

View Full Version : My learning experience developing in UE4



SwiftIllusion
05-15-2014, 07:07 PM
I felt that while I hope to bring this to OUYA, I haven't had experience yet producing Android projects from UE4, and as this is a relatively small testing project it may ever only be released as a side-loading .APK, hence why I'm not posting this in upcoming OUYA games. I also felt as this is a learning experience it may be appropriate to share updates here in OUYA developer talk.


----------------- Design Overview ==

An identifiable rough idea of the projects concept : Smash Bros. / Bomber Man / Robot Wars
Note#1: The design below may be revised and altered while playtesting and experimenting with UE4 development.
Note#2: This is intended as a small scale project to experiment with UE4, so there won't be extended development times for better art, etc.

Controls:
--Keyboard
Movement = W/A/S/D
Jump = Spacebar (hold for higher jump)
Attack 1 = left click (hold to launch, then release when you want the bomb to explode-allows strategic timing - will explode automatically after 3 seconds)
Attack 2 = right click (hold for stronger power)--Controller (OUYA/plugged in xbox360 controller)
Movement = Left Joystick
Jump = (O/A)
Attack 1 = L1
Attack 2 = R1
Combat - (VS mode)
--The arena
Battles will be fought across multiple platforms (each platform will have a randomly chosen hazard theme to intensify the arena).
You start with 3 stamina blocks which recover over time.
--Your attacks
Attack 1 will launch an explosive bomb consuming the value of a single stamina bar, deals damage to your opponents and weakens them.
Attack 2 will fire a short range lazer that grows and absorbs stamina for the period it is held, will knock opponents further than bombs.
Above your stamina bar will be your morale, which effects your resilience to the force of attacks and will swing back and forth with
your opponents in a 'tug-of-war' fashion.
Defeating opponents - (VS mode)
When there is one opponent left in the arena they will gain a round point. You can choose first to 2/4/10 points,
and whether to randomize the arena each round.]

Miscellaneous
Online multiplayer is planned, local multiplayer will be experimented with to see if it's possible
(I need another controller before I can test local multiplayer properly).
After a VS mode is developed I plan for a basic co-op mode to experiment with AI (enemies will be blocks/shapes)

----------------- Latest Video Update ==


https://www.youtube.com/watch?v=qqv5gToyg8o

Previous videos--
https://www.youtube.com/watch?v=q2_5CkyCzrU
https://www.youtube.com/watch?v=IecGTKCBvXE
https://www.youtube.com/watch?v=4V9vpMF6JCs

----------------- Latest Blog Log ==

note: I have nothing against people using Unity, however I've had nothing but awful experiences with it, yet favorable experiences with UE4, hence the tone of these Blog Logs.

UE4 vs Unity+Playmaker : Character object experience. (http://www.midnight-tinkering.com/forum/blog.php?u=2&b=27)

Previous Blog Logs--
UE4 experience update, learning Unity/Playmaker vs UE4/blueprint (http://www.midnight-tinkering.com/forum/blog.php?u=2&b=26)
UE4 test project - 'online bomberman meets super smash bros.' (http://www.midnight-tinkering.com/forum/blog.php?u=2&b=25)
Unreal Engine 4 completes me / Comparing experiences with Unity (http://www.midnight-tinkering.com/forum/blog.php?u=2&b=24)

DrunkPunk
05-16-2014, 06:55 AM
Preach on Sir Squirrel! :D

Loved the new video. Looks like things are shaping up quite nicely.

SwiftIllusion
05-16-2014, 07:27 AM
Preach on Sir Squirrel! :D

Loved the new video. Looks like things are shaping up quite nicely.
Thanks a lot DP ^^!
Also got a more concrete idea for the control scheme and design which I'll be writing up tonight with hopefully another video to share :).

DrunkPunk
05-16-2014, 07:42 AM
Thanks a lot DP ^^!
Also got a more concrete idea for the control scheme and design which I'll be writing up tonight with hopefully another video to share :).

Yay for sharing! :D

Steven7
05-16-2014, 07:51 AM
looking good Swift looking real good, glad things are coming together for you (:

SwiftIllusion
05-16-2014, 08:58 PM
Yay for sharing! :D
^^!!
Got a bit stuck trying to implement the new mechanics for the characters shooting so there's no video, but updated the games design in the original post :)

looking good Swift looking real good, glad things are coming together for you (:
Glad you're liking it :), thank you very much ^^.


Updated the first post to hopefully give you a better idea of this projects direction.
Currently trying to re-work the characters shooting to function in the multiplayer environment and with the newly described mechanics
The next video update will likely be when I can show the characters core desired combat mechanics (without a HUD yet).
Hope you like the sound of it ^^;

Killswitch
05-16-2014, 10:13 PM
Interesting...not converting at the moment but interesting.

Sitting Fox
05-16-2014, 11:05 PM
UE4 = Unreal Engine 4
For anyone else who had to stop and think a moment.

---------

SwiftIllusion,

Thank you for sharing your experiences with Unreal Engine 4. May it help those it's best suited for more easily find it.

I feel you get a little carried away with how much worse you feel Unity is in comparison at times, though the disclaimer is a nice counter on your part to pretty much acknowledge that.

To each their own. I don't use either engines myself. However, you've given some very interesting information to keep in mind should I ever decide to go down that route.

SwiftIllusion
05-17-2014, 11:42 PM
Interesting...not converting at the moment but interesting.
No worries, I understand that there are people still invested in Unity, just sharing my personal opinions and experience.

Thank you for sharing your experiences with Unreal Engine 4. May it help those it's best suited for more easily find it.

I feel you get a little carried away with how much worse you feel Unity is in comparison at times, though the disclaimer is a nice counter on your part to pretty much acknowledge that.

To each their own. I don't use either engines myself. However, you've given some very interesting information to keep in mind should I ever decide to go down that route.
No worries :)! I never thought I'd get to see someone else come out with a game engine like Unreal Engine 4, especially with it providing full access to the source code literally enabling you to do anything you want with it. And ontop of that it's actually been used by them for their own projects so as users we will benefit not only from their continued development but the improvements to the engine they find themselves needing. This is an engine I feel genuinely deserves to be successful.
My single concern is how the opening of their marketplace will affect their desire to work on core features, because Unity hasn't cared at all about adding features already included in the marketplace and prefers to sit back and profit from the addons. But not only as mentioned they are using for their own projects, they've already posted and are maintaining an informative page showing their development pipepline and schedule, where they also allow users to vote on the features they want most to direct the schedule priorities.
And obviously the price is incredible and beyond fair.

Yeah, I acknowledge the fact that I've never had a positive experience with Unity throughout the years I've used it which can make it feel biased, so I wanted to make that a clear statement up front as I realize that there are people who have had better experiences with it and still want to continue using it. I'm never trying to over-exaggerate things, it's just only being a designer I have to rely on a game engine to be able to produce my games, and I've had far too many headaches in Unity which until now was my only available option, so the frustrating times with it still feel fresh for me but I try to detail differences objectively where possible. And following Unreal Engine on twitter you can also see many other projects moving to UE4 with graphical comparisons if that's important to you.

Glad to hear I've at least given you food for thought :). If you ever find yourself more interested, UE4 allows you to pay for the first subscription and then cancel, allowing you to make games with the engine for just one months pay (however you won't get updates-and you'll still have to follow their agreement of 5% of profits, etc as far as I know).
You can also see their roadmap here - https://trello.com/b/gHooNW9I/ue4-roadmap
They also have great templates and free projects you can download and take apart to learn from :).




Still struggling with setting up the character for multiplayer, not sure if I'm using events correctly of if there's a bug or what's up but I'll post whenever there's any progress to share.

Killswitch
05-18-2014, 03:31 AM
No worries, I understand that there are people still invested in Unity, just sharing my personal opinions and experience.

You sure? You kind of look like a salesman for them.
Pull up your sleeves and prove you don't have Unreal tattoos :p
Unity, just like Unreal is fixing some of their brokenness so maybe now it will become a true race and we'll see them both improve their engines.

SwiftIllusion
05-18-2014, 05:59 PM
You sure? You kind of look like a salesman for them.
Pull up your sleeves and prove you don't have Unreal tattoos :p
Unity, just like Unreal is fixing some of their brokenness so maybe now it will become a true race and we'll see them both improve their engines.
Haha I'd never get a tattoo so you don't have to think that xD. I'm also definitely not wearing an Unreal badge either...

Well the stronger the competition the further the end result can end up becoming, so hopefully it'll benefit both of us :).

DrunkPunk
05-19-2014, 07:25 AM
Haha I'd never get a tattoo

I will have to have my friend in Oz get you drunk and tattoo you in your sleep. ;)

SwiftIllusion
05-22-2014, 11:35 PM
I will have to have my friend in Oz get you drunk and tattoo you in your sleep. ;)
Rofl, thanks but no thanks xD!


Was able to get a second controller today ^^! So will be ready for co-op testing when I get to that point.

Dun dun dunnnnnnnnn *dramatic music*.
Looks like a major issue has been discovered with UE4's character component.
But first I just want to address that updates have been slow because I've had a bunch of 'IRL' stuff going on, and have been dealing with the learning curve of UE4 which I'll go into more depth in a Blog Log when I'm ready to post another update on that.

Anyway, UE4's character component. Regrettably as great as it is I've discovered a flaw with a particular part of it, jumping. I've gone through every method of building a more flexible jump system and double jump system but it is completely broken in multiplayer. It seems that rather than synchronizing the character objects position/rotation it's taking control over the character component itself on the server and so it overrides all these jumping methods when you're on a client. I have no idea why they chose to do it that way and it has had me going crazy.
I'm now in a position where I'm waiting for a response from a question about it, and unless there is a method I haven't actually tried to get a better jump system working, I'm going to have to build the entire character control method from scratch which will take a deal of time I had hoped I wouldn't have to waste, especially if I want as much control as their character component.

To some more positive news though, after rebuilding the method of shooting bombs multiple times I've managed to get that working :)!
It has been a bit rough because I'm in that position where it's still hard to know what's a bug and what am I simply doing wrong, and trying to workout how to send the necessary data everywhere and how it acts through the server. But I've been able to get a better grip on it now so it has been going well ^^!
Just going to be working on the second attack visual and action now while waiting for a response on the jump issue.

SwiftIllusion
05-26-2014, 06:31 AM
Wanted to update on what I'm planning to do.
After deliberating about whether to try and build my own character system or not in UE4 I figured rather than do that I'll hold off on this project until they add some more of the systems they were planning (I found out they're also hoping to have physics systems and such work with the character so it will be easier for me to have people blown about by the bomb at a later date, etc).
Also what affect my decision was the 'fail faster (https://www.youtube.com/watch?v=rDjrOaoHz9s)' video by extra credits and this discussion (https://www.youtube.com/watch?v=ISutk1mauPM) on prototypes by Johnathan Blow.

Basically to try and hopefully also resolve my motivation problem and instead of losing time on the current project I'm planning to make a quick rough prototype of this new game design before fleshing out more details, that way I'll be more easily able to gauge peoples reaction to it and instead of having a book for people to read they can experience a compressed version of the game so it'll be easier also for people to hopefully anticipate the development of the project more.
Basically it'll hopefully be better for everyone.

EDIT: After a long conversation in chat I may be developing a prototype for something else instead, though the systems and design are more secret so I don't wish to express them openly here as it's part of what I consider my defining future project. Will update when details are discussed and thought out further.

SwiftIllusion
05-31-2014, 07:52 PM
Starting work on a mini-prototype-game featuring our character artists main characters from her comic (http://thebigbadwolf.smackjeeves.com/comics/1772260/all-out-cover/) in a super cute chibi form she drew them in, and a unique movement control system.
Or at least that's the goal-it's an experiment both to see if it can be properly implemented in Unreal Engine 4 and to get a feel for a unique movement system in a compressed form.

The characters: Alex / Dexter
http://puu.sh/99gkR/d66b70ab26.png

Dexter WIP
http://puu.sh/99gmI/dd8091883e.png

Mini-Game Prototype Gameplay Overview
You'll be using the unique movement control system to dodge various obstacles, e.g. lazors/arrows/swinging axes/etc while you play as Dexter trying to save Alex.

SwiftIllusion
06-03-2014, 09:20 AM
Regrettably wasn't happy how those characters were turning out, wasn't able to do them justice in 3d, they just didn't translate as well as I initially expected.

So I restarted today working with a creature design I purchased from our character artist instead, which so far I'm a lot happier with :).
http://puu.sh/9cZ3w/f512985159.png
(The tail may be shrunk down as at that size I don't believe the motion system will be able to function)

DrunkPunk
06-03-2014, 09:31 AM
That looks fun Squirrel. Reminds me of the PetitMage in FF3!

SwiftIllusion
06-03-2014, 06:52 PM
That looks fun Squirrel. Reminds me of the PetitMage in FF3!
Really happy to hear that ^^. Have tweaked it a bunch tonight and before I add the tail I'll be texturing the rest tomorrow :).
*googles* Ahh yeah I see what you mean ^^.

SwiftIllusion
06-06-2014, 06:14 PM
Spent today having fun with the new vehicle system added to Unreal Engine 4.2....it's epic!

https://www.youtube.com/watch?v=gQVZrwOAnpw
Video Description-

I spent today playing about with the new vehicle system added to UE4, and it was all kinds of awesome. This is a video of the final vehicle settings I was happy with for more enjoyable fun feel.

It's an incredible feeling to have such a deep vehicle system readily available, I can't wait for future projects involving vehicles now :)!
If there was one thing I felt missing it was a good Boost/Nitros system built in. Here to try and emulate it to a degree I setup just 2 gears for the vehicle, and have holding shift down trigger the second, faster gear, while releasing it returns you to the lowest gear.
Still it's an incredible headstart and not everyone needs nitros/boost for their car. Short of that I can't see anything missing.

I honestly believe the default settings in the vehicle template they provide you feel better than some cars in AAA commercial games.
It's really awesome to have that in there :D (Already thinking of how good it'll be with the Pirate Land Boat vehicle :D)

----

Anyways now I've had fun with that, back to working on the prototype at hand :).

SwiftIllusion
06-07-2014, 08:38 PM
Was caught up helping my mum today, buuut I GOT THE TEXTURE FINISHED :D.
Hope you like it :) (not shaded in-game yet but this closely resembles it)

http://i.imgur.com/4NLo7YP.png

Working on the tail next :), and it will be at full size so I can best test the limits of the motion system prototype and see if it's able to function as I need it to.

SwiftIllusion
06-09-2014, 08:37 AM
Finished the model/texture :). Animation next. Hope you like the final result, I'm super happy with it ^^;

http://i.imgur.com/rAxRrpq.png

DrunkPunk
06-09-2014, 09:01 AM
Looking good!

SwiftIllusion
07-01-2014, 09:39 PM
Looking good!
Thank you ^^;!


Bunch of IRL stuff and thoughts and procrastinations about how to move forward, etc, was cause for some delay but I've been pushing forward on this project and I know where I want to take it now, but first I'd really like to hear your thoughts on it so far ^^;

WIP video demonstrating the early stations of the motion system, everything's controlled with just the L1 and left joystick (After working on this in UE4 I can say it allows all my advanced motion ideas to be brought to life, and damn I jumped about as happy as I could be the first time I had my character moving the way I wanted it to :D)


https://www.youtube.com/watch?v=dkNORqD07CI

DrunkPunk
07-02-2014, 07:09 AM
That video made me feel like someone had roofied my drink...very trippy and I have no idea what was going on, haha!

All the movements are looking good though, and of course your character is super cute.

SwiftIllusion
07-02-2014, 08:20 AM
That video made me feel like someone had roofied my drink...very trippy and I have no idea what was going on, haha!

All the movements are looking good though, and of course your character is super cute.
XD!

Thanks :). It feels amazing to control so I'm super happy with it :D.



Current plan is depending on how progress on it is received to make it a rogue-lite-randomly generated rooms/traps treasure heist adventure, where while there is no combat your tasked with dodging the traps effectively with the character to capture some kind of treasure. Difficulty settings will tweak the level of slow motion/speed your character can react, etc so that players of various twitch-reflex-ish ability can enjoy a tough challenge. Those drones were mostly stand-ins to demonstrate the character and test working on AI.
The overall positive is that it's given me more experience with various areas of UE4 and it's slowly becoming more natural to work in and knowing how to achieve the results I want.
The biggest current design-limiting factor is they're still working on the Visual Scripting GUI system so what can be done is really limited.
While that doesn't negatively effect this projects design plans it's one of the reasons why (ontop of really liking this design) I'll likely continue expanding this until the system is available so that I can move smoothly back to the Torzig project, which needs more advanced GUI.

On a UE4 orientated side-note, this is something they made for demonstrating the mobile nvidia K1 GPU - https://www.youtube.com/watch?v=X-tAZtbDZ8E

SwiftIllusion
07-04-2014, 07:22 PM
Sharing my most recent 'Blog Log' here even though it's somewhat more personal/not entirely UE4/project related.

"Many thoughts feelings and discussions as an indie game designer"
I feel that with the tone of this more Blog-ish 'Blog Log' I must ask you to understand, this isn't intended to sound like a 'woe is me indie developer please feel sorry for me and give me all your monies' (especially when I'm still currently making game and distributing them for free), I'm completely aware of the various levels of success developers have and there are surely people better and worse off than myself, etc etc. I'm just here writing about my own feelings, thoughts, and discussing various areas of indie gaming based on my own experiences in comparison to those I'm most often exposed to.


Struggles with Motivation
Something I find myself struggling with consistently when working alone on developing projects is motivation.
(I commission concept art/music from some incredibly awesome people-though I work alone in that I design/develop/test/play/etc the project myself).
While it's not impossible for me to be motivated, I more often find myself in a situation where I'm overcome by concerns about how to move forward, or I feel that it's not good enough for me alone to enjoy the game I'm working on and without larger reception or positive consistent feedback I find it too easy to design a new game and see the reception to that instead.
The time I felt most motivated was when I had the incredible honor of designing and developing Bimey alongside their team.
I was so excited to start on the project after getting home from work and reading their email I honestly put my shirt on backwards and didn't find out until my Mum told me xD;
It was an incredible experience because I was able to hear directly from them feedback at each stage and how much they were enjoying it, and because my overall motivation for game design/development itself is to bring people joyful experiences to players I was consistently able to enjoy making progress with it and focusing on it completely without doubt clouding my mind.
Now though I'm once again developing these projects by myself without someone I can consistently discuss the project with or playtest the experience with so it's hard to force myself to be as motivated as I would like to be.
However now while I still have the inherent fear and doubt regarding the current project I've grounded myself in the reality that I must complete it and hope that the final resulting experience will be something people can enjoy and share, rather than backing away once again and presenting another design.

Dreams, Fears and attempts to find the right path forward
Another reason I have so often decided to present a new game design and hope for a larger response instead of moving forward with each one I personally enjoy is because of my dreams and wishes for the future.
Without being melodramatic and going into all the details my only path forward in life is game design, it's the single area I've found talent in and it isn't reasonable to believe I'll be capable of holding any other job in the future. Because of this I can't simply leave this as a hobby of mine, I can't find comfort in people saying 'oh well you've got all the time in the world', etc. I need to be able to move forward, I need a game that can spread and people can enjoy and develop a following who look forward to future projects.
Ontop of this I'm in a position where I've had the honor of meeting some incredible individuals, both in talent and personality. I wish to be able to continue working with them on much larger scale designs I have in mind when I have the resources necessary to afford such projects, because for me it's not as simple as finding a 'talented' team, but rather building a team of awesome people I can have faith in. They are all both awesome and extremely uniquely talented, and my largest fear is that I won't reach a position where I can work with them on larger scale projects before they lose interest or find better opportunities I couldn't provide myself in time.
This is arguably my largest reason for too often presenting a new design and changing projects in the hopes that it may gain larger traction, because I want to reach a large audience and level of success as soon as possible so that I don't end up losing a possible future with such awesome people, as even now I couldn't imagine how I would move forward without them, they are irreplaceable and every moment with them is an honor.

Problems with being strictly a game designer trying to gain traction on projects
I'd like to address here what I feel makes finding traction on a project virtually impossible as strictly a designer in the indie game industry.
Being just a game designer, I've basically only got 'programmer art' to present my projects before completion, except I can't program.
This means that the common methods of gaining traction or acknowledgement via things like project screenshots/trailers are impossible as I'm not an artist, and participating in Ludem Dare competitions etc is impossible as I'm not a programmer. Additionally I have an awful time writing or talking as I have a frustratingly large amount of difficulty translating my thoughts so I end up hating how childish/unintelligent I sound and this limits me from things like dev diary's or intelligent posts which have helped some projects gain momentum.
So how do I try and present my projects and gain traction?
Previously I've written out extensive designs for projects along with summarized points, etc, as that's the only thing I can do as a designer. The issue here is that most people don't want to read that much anymore (have responded to questions that were already answered in the summary), and for those who do read it many still won't have confidence in it as I don't have any large successful projects to present.
This is why I commonly find myself in a position with a lack of motivation and attempting to resolve the issue by presenting a new design as designing new games are easy for me and that's all I felt I could do. However with a lot of recent thoughts I became more aware of this issue, or rather its specifics became more clear, which is why rather than letting myself fall into doubt once again with the lack of early traction and presenting a new design, I am resolving myself to continue with this latest project until its completion so that I can present a unique experience that may hopefully be shared and enjoyed, and rather than rely on attempts to gain momentum towards the future during the early stages of a new project, instead gain it via the completion of a successful project.
Regarding this project and updates, I plan to wait a great deal longer between updating the topic here,
however if you wish to follow progress I plan to start back up these blog logs which you can view here - http://www.midnight-tinkering.com/forum/blog.php?u=2

DrunkPunk
07-07-2014, 09:06 AM
Long-form Squirrelage done proper there.

As always, a good read. Thanks for sharing! :D

Killswitch
07-07-2014, 11:25 AM
Sir Squirrel,

Developer logs are for ramblings and thoughts. You seem to come off fine in most cases so this is a case of YOU WORRY TOO DAMNED MUCH.

-Mr 'Switch :D

SwiftIllusion
07-07-2014, 04:06 PM
Long-form Squirrelage done proper there.

As always, a good read. Thanks for sharing! :D
haha XD!!

^^! Thank you very much, happy to hear you valued the time spent reading it :).

Sir Squirrel,

Developer logs are for ramblings and thoughts. You seem to come off fine in most cases so this is a case of YOU WORRY TOO DAMNED MUCH.

-Mr 'Switch :D
Haha, well you know, it's the internet. Too often I feel like I'm in the middle of a pack of lions hoping they've already eaten or they won't be hungry soon. Always fear if I let my guard down I'll be attacked unnecessarily xD;

Thanks for reading it :)

Killswitch
07-07-2014, 09:44 PM
haha XD!!

^^! Thank you very much, happy to hear you valued the time spent reading it :).

Haha, well you know, it's the internet. Too often I feel like I'm in the middle of a pack of lions hoping they've already eaten or they won't be hungry soon. Always fear if I let my guard down I'll be attacked unnecessarily xD;

Thanks for reading it :)

Well the pack of lions part is right.....

DrunkPunk
07-08-2014, 06:19 AM
Well the pack of lions part is right.....

Rawr and stuff.

HeadClot
07-16-2014, 04:06 AM
Very Cool!

Another Unreal Dev on the OUYA forums :)

SwiftIllusion
07-17-2014, 02:25 AM
Very Cool!

Another Unreal Dev on the OUYA forums :)
Yup ^^.



Some of the pure insane epicness added in Unreal Engine 4.3.


Render 3D Imposter Sprites

Imposters are like sprites that are represented from all view angles so they can rotate and light in 3D.
https://de45xmedrsdbp.cloudfront.net/blog/43_PreviewImages/3dImposterSprites-1275x621-767943032.png

Example Imposters:
One of the meshes near the middle is a real static mesh, the rest are imposters.
https://de45xmedrsdbp.cloudfront.net/blog/43_PreviewImages/ExampleImposters-826x408-1167069668.png

It also supports a mode called "Single Rotation Axis" which is meant to represent a billboard for a distant object on the horizon. What that does is only rotate the mesh around Z so that you get either more frames or more resolution than a full imposter.
https://de45xmedrsdbp.cloudfront.net/blog/43_PreviewImages/SingleRotationAxis-635x622-235500026.png

This is a test scene using the Single Rotation axis imposters. Here the material was set to randomly rotate the position of the 'flipbook' by using worldposition, giving each a unique rotation.
https://de45xmedrsdbp.cloudfront.net/blog/43_PreviewImages/WorldPosition-752x289-542109789.png

NOTE: above, imposters are A SINGLE TEXTURED PLANE!!!
For example a previous project on OUYA where I had to limit the amount of foliage greatly for performance, not only could I now increase the detail of the foliage but have much, much more of it as well! That's just beyond incredible :D!! Of course there are cases where it might not suit but for something like ground foliage it's incredible :D.
Here's a video demonstrating it a bit more at 18:40 (https://www.youtube.com/watch?list=PLZlv_N0_O1gbggHiwNP2JBXGeD2h12tbB&v=X0sHyv6pvgw&feature=player_detailpage#t=1120)

SwiftIllusion
07-17-2014, 09:58 PM
Our artist Robin Keijzer (http://robinkeijzer.deviantart.com/) just finished the commission for designs for this project, and I'm insanely happy to say the least :D,
hope you all like it too :).

(click image for full size)
http://i.imgur.com/HOsvdEal.jpg (http://i.imgur.com/HOsvdEa.jpg)

SwiftIllusion
07-18-2014, 06:00 PM
First results up, the room :)!

http://i.imgur.com/cPCbRP3.png
http://i.imgur.com/YslqrvR.png
http://i.imgur.com/zMI2gfZ.png

SwiftIllusion
07-20-2014, 03:08 AM
Adjusted ground graphics and first enemy.
http://i.imgur.com/MHKEN3m.png
http://i.imgur.com/Ibvqoyb.png
NOTE: This is all unlit with a shader optimized for the possible OUYA build.

And here is the rough development pathway
> Develop new enemy and get it completely functional within the new environment (and implement difficulty logic in AI).
*show video of new enemy in-game
> Re-animate character with reworked tail rig (found large issues when trying to begin animating the backflip), and add full movement functionality refining all the blueprints and add self-contained gameplay logic (e.g. slowmotion cooldown).
*show video of fleshed out character motion system
> Build all reaction logic between player and enemy (receiving damage/death)
> Build self-contained room logic (enemies spawning on room load, difficulty effecting enemy spawning)
> Setup the dynamic layered music tracks to react to the player and enemies actions.
*show video of a couple rounds with the new systems and music
> Build 2 new enemies and add them to the room spawn logic.
> Expand room logic to allow exiting room after surviving for a time and spawning the next room.
*more lengthy playtest to refine all systems and squish any bugs that might pop up
*show video of a couple rounds with the new systems
*prepare a first alpha build if interest is shown for a preview
> Listen to feedback and tweak difficulty levels based on response.
> Add 2 traps and expand room logic for navigating through a room within an amount of time.
*show video of a couple rounds with new traps
> Expand room logic for different floors, stages, and have these details represented in the room.
> Add treasure chest for final room.
> Build menu and add statics for end-game screen, etc.
> Complete the remaining enemies.
> Touch up all remaining areas of the game and do a lengthy playtest (possibly stream it, thanks to UE4's awesome twitch integration :)?)
>>> Release game and hide in a corner until I'm brave enough to see if there's a response to it.

Jawdy
07-20-2014, 12:39 PM
Swift - have you deployed to the OUYA yet?
I'm just wondering as we're experimenting with other engines and technologies and, back from my UDK/UE3Mod days, I've thought about joining the UE4 bandwagon :-) But, with a little research, the OUYA's not on the official list, so I was just wondering what sort of performance you might be getting out of it, when deployed to the little cube?

Jawdy
07-20-2014, 02:53 PM
OK, so my impatience got the better of me and I signed up, have the UE4 launcher (and 4.3 Engine version)... fired up the tutorial documents, started a "Blueprint" FPS base project and quick-deployed to the OUYA (didn't change any settings).

It's, apparently, running at 1080p, definitely not at above 25fps, but this is a great start! The OUYA controller works out the box, which is fantastic as I'd read it didn't on previous versions... so now it's a case of tweaking the parameters, dropping the res' to 720p (as there's still the issue with fill-rate on these older Tegra devices) and hopefully get it rockin' above 30fps or even closer to 60!
Granted, this is without any effects, everything baked and simply a test of the hardware with the engine, but wow! The editor has been overhauled since I last used it (love the browser-like tabbing :-) ), so this might be a great way of making games for the OUYA! :-D

SwiftIllusion
07-20-2014, 03:16 PM
Swift - have you deployed to the OUYA yet?
I'm just wondering as we're experimenting with other engines and technologies and, back from my UDK/UE3Mod days, I've thought about joining the UE4 bandwagon :-) But, with a little research, the OUYA's not on the official list, so I was just wondering what sort of performance you might be getting out of it, when deployed to the little cube?

Haven't tried deploying to the OUYA yet because I hadn't got the project anywhere significant yet, and after doing a bunch of searching while there's an ODK for UDK there's nothing for UE4 yet, additionally saw some input issues with the only possible suggestion basically being 'program it yourself' but as a non-programmer that's of no help at all. So not looking forward to trying to get it onto the OUYA, though I plan to attempt it once the new enemy is completed.
Performance wise I know they've been consistently working on improving mobile functionality/performance and there's a bunch of good looking free examples you can download that have been made specifically for mobile. If you'd like to try it out yourself you can just pay $19 for the subscription then immediately cancel it-you just won't have access to updates.(you'll still need to pay 5% above 3k earnings though). Regardless though I'll definitely report on my attempts to export to it when I get the chance to.

And lmao, just checked to see if there were responses before posting and looks like you beat me to it and had success I hadn't dreamed of based on the results I'd found, awesome to hear it working so significantly even if you're experiencing low framerate :D. Thanks a ton, now I can at least breath a little easier when I try it out myself, will be sure to report on the FPS for my project when I do :)!
And yes it's definitely a whole lot of awesome, happy to hear you enjoying your early experience with it :).
In-case you aren't aware they've got a ton of epic official UE4 tutorials over on their youtube channel - https://www.youtube.com/playlist?list=PLZlv_N0_O1gaCL2XjKluO7N2Pmmw9pvhE
And you can check out their planned Roadmap of development here - https://trello.com/b/gHooNW9I/ue4-roadmap

Jawdy
07-20-2014, 03:33 PM
Haven't tried deploying to the OUYA yet because I hadn't got the project anywhere significant yet, and after doing a bunch of searching while there's an ODK for UDK there's nothing for UE4 yet, additionally saw some input issues with the only possible suggestion basically being 'program it yourself' but as a non-programmer that's of no help at all. So not looking forward to trying to get it onto the OUYA, though I plan to attempt it once the new enemy is completed.
Performance wise I know they've been consistently working on improving mobile functionality/performance and there's a bunch of good looking free examples you can download that have been made specifically for mobile. If you'd like to try it out yourself you can just pay $19 for the subscription then immediately cancel it-you just won't have access to updates.(you'll still need to pay 5% above 3k earnings though). Regardless though I'll definitely report on my attempts to export to it when I get the chance to.

And lmao, just checked to see if there were responses before posting and looks like you beat me to it and had success I hadn't dreamed of based on the results I'd found, awesome to hear it working so significantly even if you're experiencing low framerate :D. Thanks a ton, now I can at least breath a little easier when I try it out myself, will be sure to report on the FPS for my project when I do :)!
And yes it's definitely a whole lot of awesome, happy to hear you enjoying your early experience with it :).
In-case you aren't aware they've got a ton of epic official UE4 tutorials over on their youtube channel - https://www.youtube.com/playlist?list=PLZlv_N0_O1gaCL2XjKluO7N2Pmmw9pvhE
And you can check out their planned Roadmap of development here - https://trello.com/b/gHooNW9I/ue4-roadmap

Right now I have a quick Blueprint in place to make some console calls in order to drop the res and show the frame rate - down at 720p, frame rate was coming in at just under 10fps, with the "Blueprint FPS Demo". Trying "1024x576", just to see if I can boost the performance a little. I've already dropped the Quick Settings Engine Scalability and Material Quality Settings to lowest, but that doesn't seem to solve it.

Hmm, right now I don't think it's listening to the r.SetRes call... might need to have that happen somewhere else!?

For our needs, this might be a no-go... which is a shame, as I'm already finding the UI/Editor a joy to use, the community has answers for almost everything I'm searching for and Epic's own YouTube channel (thanks for the link! They have the same channel link show-up once you register :-) ) covers so much ground to get you started. Just need to try and squeeze out 30+fps and then it'll be usable, just not as it stands.

SwiftIllusion
07-20-2014, 04:55 PM
Right now I have a quick Blueprint in place to make some console calls in order to drop the res and show the frame rate - down at 720p, frame rate was coming in at just under 10fps, with the "Blueprint FPS Demo". Trying "1024x576", just to see if I can boost the performance a little. I've already dropped the Quick Settings Engine Scalability and Material Quality Settings to lowest, but that doesn't seem to solve it.

Hmm, right now I don't think it's listening to the r.SetRes call... might need to have that happen somewhere else!?

For our needs, this might be a no-go... which is a shame, as I'm already finding the UI/Editor a joy to use, the community has answers for almost everything I'm searching for and Epic's own YouTube channel (thanks for the link! They have the same channel link show-up once you register :-) ) covers so much ground to get you started. Just need to try and squeeze out 30+fps and then it'll be usable, just not as it stands.
I can't imagine how you're getting such a low framerate, I've heard how many high and low end mobiles there testing there pathway with in their livestreams and I couldn't see it being unplayable like that. Maybe try and download one of the mobile examples from the library and build that to the OUYA? There may be something in the blueprint FPS demo you're missing. Have you deleted the lights from the scene? Have you adjusted "[SystemSettings] r.setres=1280x720" in the DefaultEngine.ini?

Jawdy
07-20-2014, 06:05 PM
I can't imagine how you're getting such a low framerate, I've heard how many high and low end mobiles there testing there pathway with in their livestreams and I couldn't see it being unplayable like that. Maybe try and download one of the mobile examples from the library and build that to the OUYA? There may be something in the blueprint FPS demo you're missing. Have you deleted the lights from the scene? Have you adjusted "[SystemSettings] r.setres=1280x720" in the DefaultEngine.ini?

It's very odd!
First thing was to disable Mobile HDR in the project settings. So that's off - and from what I gather, Tegra 3 wouldn't even launch if that was enabled.
I tried a bunch of the test projects that come with the Editor. The simple 2D project ~50fps. But the FPS and Ball Rolling demos, <9fps.

I tried attaching r.SetRes to a Console command on launch through Blueprint.
And, after that had no effect on mobile, have tried the [SystemSettings] tag, but that made no difference on either Windows or Mobile.

I've tried dropping the Scale to 10 (the lowest value) and with an incredibly simple scene: one light (baked, no dynamics); basic plane for the "ground; one "flying" actor. I removed the skybox, just to see if that would help and from there, can get an improvement... some 30+fps if flying with the default plane taking up around 75% or less of the view, but dropping to 18fps if you're at the "plane level".

From the "Marketplace" in the Launcher, there's currently only one Mobile demo and it says it doesn't yet fully have Android support... plus, with the OUYA having somewhat dated hardware (by UE4s standards!), and not on their actually supported list, I doubt the high-end Mobile demos would function... or function well.

It really is a shame and I don't want to seem like I'm slamming the little cube, as most on here know, I'm a huge fan and advocate! I would have just liked to use an industry-standard tool for much of our development.

Swift: If you manage to find some demos or render settings that can get something playable, PLEASE let me know! I've had <1 day of playing with the engine and I must say, I'm absolutely impressed! If I was building pure PC games, I'd be 100% sold and if greater support for legacy devices (or at least, an even greater ability to turn off the more advanced features!) comes into play, then it would be a no brainer!

Good luck with your game and I'd wager you'll find a way to get all the performance out of the OUYA, but for now (for me), I just can't afford anymore time.

SwiftIllusion
07-20-2014, 06:25 PM
Trying to get a test running now but it's taking forever because I have to set it all up again (haven't built anything since upgrading to 64 bit system so didn't even have android sdk installed, let alone all the usb/OUYA malarkey setup I've long since forgot about). I'll need to sleep soon though hopefully I'll have all this working before then, otherwise I'll keep working on it and report tomorrow, having spent time with the engine I'm happy to hear you enjoying it so much as I have also and likewise couldn't bare to leave it alone now, it does too much right and just keeps getting better with each update.

HeadClot
07-21-2014, 02:50 AM
This should help getting things up and running with the OUYA.

https://forums.unrealengine.com/showthread.php?14430-Steps-to-make-your-unreal-game-work-on-ouya-(tested)

SwiftIllusion
07-21-2014, 07:48 AM
This should help getting things up and running with the OUYA.

https://forums.unrealengine.com/showthread.php?14430-Steps-to-make-your-unreal-game-work-on-ouya-(tested)
I eventually got it working but didn't have to do any of that (the controls worked perfectly), though that might have been thanks to their latest update, thanks for the link regardless :).


Also Jawdy, got my project up and running in there with everything working and 9 clones of the enemy at 30fps :)!!,
have been doing a bunch of testing though with the mobile temple project. So far nothing I have done has changed it, it's still at only 4fps which is crazy,
though it looks to have a lot of high level effects going with reflections and everything so I'm trying to see if I can find the big fps killer in there.

Killswitch
07-21-2014, 08:35 AM
4fps? Damn...

Jawdy
07-21-2014, 10:13 AM
This should help getting things up and running with the OUYA.

https://forums.unrealengine.com/showthread.php?14430-Steps-to-make-your-unreal-game-work-on-ouya-(tested)

That sorts the 2D side of the engine, but not the 3D side... and I didn't have to do any of that! Just used version 4.3, hit deploy and the sticks and buttons all work :-)


I eventually got it working but didn't have to do any of that (the controls worked perfectly), though that might have been thanks to their latest update, thanks for the link regardless :).


Also Jawdy, got my project up and running in there with everything working and 9 clones of the enemy at 30fps :)!!,
have been doing a bunch of testing though with the mobile temple project. So far nothing I have done has changed it, it's still at only 4fps which is crazy,
though it looks to have a lot of high level effects going with reflections and everything so I'm trying to see if I can find the big fps killer in there.

That's fantastic!!!
I'd be really interested to find out why the very basic scene I built was running between 18fps and just over 30fps, with nothing else going on!


4fps? Damn...

Yeah. I'd wager this is down to UE4s new physically based shaders. I have no doubt that this runs beautifully on Tegra 4/K1 or higher... just not on our little cube.

SwiftIllusion
07-22-2014, 09:16 AM
That's fantastic!!!
I'd be really interested to find out why the very basic scene I built was running between 18fps and just over 30fps, with nothing else going on!
Thanks ^^! I'm super happy to see it perform like that on there.

After a ton of testing it looks like the biggest effector on performance was the reflections, but it still didn't help the OUYA build.
Even removing a lot of the geometry made no real impact, so that wasn't the issue, and it shot straight to 30fps even on the OUYA when looking
at only the sky/distant islands which are both unlit. So it looks like it's regrettably likely something to do with the way the shaders function,
as based on what I read there was enough change between Tegra 3/4 in the way it worked that it's just recommended for Tegra 4/above
as Jawdy mentioned, and you can see their latest achievement with K1 here (https://www.youtube.com/watch?v=X-tAZtbDZ8E).
Though 2D games are fine because they aren't using 3d shaders, and my projects fine because I'm using unlit shaders with fresnel to get fake shading.
The best I got was around 6fps when walking around the general scene, though to be somewhat fair even on my PC it was 20-30fps
(though that may have been effected by running it in-editor).

Also in regards to the resolution, it looks like while you can set the resolution on PC/etc, for packaged projects like ones sent to OUYA,
it depends on an awesome technique via the Engine Scalability>Resolution scale, which can be read here (https://docs.unrealengine.com/latest/INT/Engine/Rendering/Scalability/ScalabilityReference/index.html).
It's likely done so you don't have to worry about the various native resolutions of mobiles, and you can get the screens resolution via blueprint
during run-time for scaling/moving about GUI (as it mentions the resolution doesn't effect the specifically GUI objects).

So in short it's awesome but prettier projects needing lighting, etc won't be able to publish a finished game until OUYA2.

SwiftIllusion
07-29-2014, 06:34 PM
The AI work has been going pretty well but ran into a block with using their path navigation system.
Though it's resolved now while I was waiting for a response in the answerhub I started work on the
next enemy as I decided it may be better to show 2 enemies for the next video rather than just 1,
so here is a screen of the reworked flying drone :).

http://i.imgur.com/08h2gkw.png

SwiftIllusion
08-02-2014, 10:35 AM
Good news engine wise, UE4 just posted some documentation of the initial experimental stage of 'UMG' (Unreal Motion Graphics/their interface solution),
and it's already sounding absolutely insanely awesome, especially the 'Viewport' widget oh my goodness.
https://forums.unrealengine.com/showthread.php?25533-UMG-Documentation-for-4-4-Preview&p=109116&viewfull=1#post109116
That has given me confidence that they'll be building a strong enough interface solution for all my future needs, can't wait to see it grow :)!

SwiftIllusion
08-22-2014, 04:20 AM
After getting super sick, migraines, IRL stuff coming up, project corrupting, finding out rotation rates were offset by framerate, and then again finding out they ignored timescale (so when I played the game at the intended speed the opponent rotated too fast), discovering the UE4 interface solution blocked input from a controller, having to give up on getting the scoring implemented with little time left before the Ludum Dare competition starts, finally have a recording of "Bunnyball" to show.


https://www.youtube.com/watch?v=q0_aDMSqQMY

Ended up spending a lot of time playtesting and tweaking the enemy AI to strengthen the single player experience, hopefully you like what you see :).
Depending on the response I may add multiplayer to it after the Ludum Dare competition, but at the very least I'll be cleaning it up for its single player experience.

DrunkPunk
08-22-2014, 06:46 AM
Sad to hear you're sick, but good to see you back to posting. I do look forward it them. :)

Fun! Like 3D Pong with flapping bunny ears style!

SwiftIllusion
08-22-2014, 08:58 AM
Sad to hear you're sick, but good to see you back to posting. I do look forward it them. :)

Fun! Like 3D Pong with flapping bunny ears style!
Thanks, I just have to cross about a hundred fingers more than I have to hope that I don't get screwed over as much during the Ludum Dare Jam, it has reached a point where it's beyond simply important for it to be successful for me, so I pretty much have no room left to feel excitement or anticipation, just gotta hope it works out.

Thanks :D! Haha yeah :)! I wish I was better at it though XD, one of the first things that made me focus so much on improving the AI, because I couldn't play well enough to show off its potential xD;

SwiftIllusion
08-28-2014, 06:28 AM
Rage: Destroyer of worlds
http://www.ludumdare.com/compo/wp-content/compo2/375043/17539-shot0.png
http://www.ludumdare.com/compo/ludum-dare-30/?action=preview&uid=17539

As written in the description, it's not as complete as I wanted it to be, however I'm hoping to round it off some more and submit it for this months '1 game a month (http://www.onegameamonth.com/)', though if I'm not happy with where I get it before then or people would like to see even more I'll likely keep going until the voting finishes and release it afterwards as a post-compo build. Regrettably though there's a UE4 bug where when using their new interface system it disables the controller input for some reason, though if you enjoy it enough to wish for an OUYA build I'll be trying to keep the control scheme something that will translate well to controller for a possible OUYA build :). And may afterwards finish up what I have of the Bunnyball for an OUYA build too and leave it as-is, as I hope to get back on track with Skratrix and aim for a release to enter next months '1GAM (http://www.onegameamonth.com/)' which is permitted in the rules.

Can I start early?
Yes. All that counts is what month you RELEASE your game on. Even if you've been working on a pet project for years, as long as you finish and launch it during onegameamonth you are warmly welcomed to include it on our site. The more the merrier.

HeadClot
08-28-2014, 10:01 PM
Optimization is key with unreal engine especially when running on the Tegra 3.


https://www.youtube.com/watch?v=HY62PAsM7eg

My advice use unreal engines forward render instead of the deferred render,

That video should cover it :)

SwiftIllusion
08-29-2014, 12:39 AM
Unreal Engine 4 doesn't actually have forward rendering - https://docs.unrealengine.com/latest/INT/Engine/Rendering/Overview/index.html
"Deferred Shading

All lights are applied deferred in Unreal Engine 4, as opposed to the forward lighting path used in Unreal Engine 3. Materials write out their attributes into the GBuffers, and lighting passes read in the per-pixel material properties and perform lighting with them. "
For me though I'm not using lighting just relatively simple shaders to get an effect so that helps a lot with optimization.

SwiftIllusion
09-01-2014, 08:14 AM
A video showing progress towards the 'OneGameAMonth' submission :).

https://www.youtube.com/watch?v=7C-zCQHYasQ

SwiftIllusion
09-04-2014, 03:21 AM
I decided rather than pulling features and rushing again for an incomplete experience to be released in time for the 'OneGameAMonth' entry, I've chosen to continue working on the project for a release after the Ludum Dare voting ends, so here's another video showing me playing with the addition of elemental traits/effects :).


https://www.youtube.com/watch?v=4QF3woJ2OdM

Here's more details on its progress if you're interested - http://www.ludumdare.com/compo/2014/09/03/rage-destroyer-of-worlds-progress-towards-post-compo-release/

Also will likely be making another large thorough post regarding experiences with working with UE4 for this after it's completed.

DrunkPunk
09-04-2014, 12:35 PM
There has never been a game titled "Rage: Destroyer of worlds!" that looked so cute... ;)

SwiftIllusion
09-15-2014, 03:19 AM
There has never been a game titled "Rage: Destroyer of worlds!" that looked so cute... ;)
Thank you :D



New gameplay video and demo available :D


https://www.youtube.com/watch?v=Wjc_z1k5G4E

I'd like to have a demo available for the OUYA but somehow the latest version of UE4 has broken exporting to android (I can export previous projects that used previous versions of the engine fine), so until this is resolved I hope those with PC would like to try out and give feedback on the latest release :).

https://www.dropbox.com/s/i3q8mm0dbv6kek9/RageDOWPostCompoDemo.zip?dl=0

---Controls (keyboard | controller)--
Movement-- w/a/s/d | Left Joystick
Camera-- Mouse | Right Joystick
Invert Vertical Camera Motion-- Middle Mouse Button | R3
Swing Attack-- Left Mouse Button | R2
Strike Attack-- Right Mouse Button | R1
Jump-- Spacebar | L1

---Gameplay Notes--
This shows your combat data indicators - http://i.imgur.com/mJTzMIY.png
This shows you what chests are available to get and how they help you - http://i.imgur.com/aRvDg1p.png
Currently there is no 'end world', however the final build is planned for you to only combat 5 worlds before facing the final enemy. It's intended to be a challenge to reach that point and your score will be based upon how many enemies you defeated.
Currently you intentionally don't die, so that you can continue forward and balance the difficulty growth of enemies throughout each world/how the stat increases balance, etc, without worrying so much about death. It's more a test of how it feels currently.

Hope you enjoy this latest release ^^

SwiftIllusion
09-19-2014, 08:30 PM
Fire and Ice are in-game and functional now, video update :D.


https://www.youtube.com/watch?v=oZe-5-nHaDU

Fire Element: This causes you to be damaged over time, while also making you run faster-but turn slower.
Ice Element: This causes you to be unable to move beyond turning/attacking. The next attack you are hit with has double the chance to deal critical damage, and frees you from the ice.

The final elements to be added are Earth and Lightning, then it's a bunch of playtesting/balancing, adding the interface and it will be complete :).

SwiftIllusion
09-23-2014, 04:34 AM
The rest of the elements have been added, and it has been tons of fun to play so far, hope you all enjoy it too :D!


https://www.youtube.com/watch?v=H7lBnGjzoqA
You can get this latest build here-it isn't final as of yet but short of the menus/end-game (intended to be stage 5) and possibly more minor tweaks it's done :) (have playtested it for multiple hours and feel the only last adjustment need is player health which I'll finalize after adding in player death).
Download for PC here - https://www.dropbox.com/s/lzk2ttaut6vza6q/RageDOWfinalbeta.zip?dl=0
(Have to workout weird performance drops before an OUYA version can be made available, so I'll do that after the final build is finished)
---Controls (keyboard | controller)--Movement-- w/a/s/d | Left Joystick
Camera-- Mouse | Right Joystick
Invert Vertical Camera Motion-- Middle Mouse Button | R3
Swing Attack-- Left Mouse Button | R2
Strike Attack-- Right Mouse Button | R1
Jump-- Spacebar | L1
---Gameplay Notes--This shows your combat data indicators – http://i.imgur.com/mJTzMIY.png
This shows you what chests are available to get and how they help you – http://i.imgur.com/aRvDg1p.png
Currently there is no ‘end-game’ or death condition for testing purposes (so you don’t need to worry about dying when experimenting), however the final build will have you successfully complete the game upon defeating the 5th worlds boss.