Help a newbie pick an engine

Friendly conversations, and everything that doesn't fit into the other forums.
23 posts Page 1 of 3
throw_away
Posts: 8
Joined: Tue Oct 04, 2016 4:27 pm
  by throw_away » Tue Oct 04, 2016 5:35 pm
Hi,

So in my quest to find an open source 3d game engine for my project i've finally decided that i need to choose between either Torque3d or Urho3d. Both seem quite capable and torque has great tools and a lot of shipped titles which is a huge plus IMO for any engine as it shows it isn't just some fancy toy. Urho does not have any shipped titles that i'm aware of and the tools are either lacking or just not very good. However, urho is pretty straightforward to program as it follows the entity-component paradigm and since i have a programmer background it feels even better because urho is very programmer-centric. Torque has torquescript, which is fine but i still don't get the whole client/server thing everywhere. For a game that is supposed to be single-player only, it's quite confusing what needs to be server-side and what needs to be client-side.

So for people who are much more familiar with the engine, how difficult is it to make a game with torque that is NOT a multiplayer first person shooter? I ask this because looking at torque, i like almost everything about it except for the way it's supposed to be scripted/programmed. Programming is a significant portion of every game but it seems kind of foolish to make a decision based on that alone since there are a lot of other important parts to making a game as well. The game i have in mind is first person with little to no shooting and mostly taking place in one persistent but small level/world. I'll be the only person working on this project but i might involve other people in later stages for better models/music etc. I've made small games/demos before in other engines and dabbled with opengl a bit as well but never shipped a commercial title which is why i still have a lot of holes in my knowledge. :P

Would you suggest sticking with torque and learning more about it and it's internals until it all "clicks" or would urho be a better option since it seems easier to get started with?

Thanks
Johxz
Posts: 447
Joined: Sat Feb 07, 2015 11:37 pm
by Johxz » Tue Oct 04, 2016 7:40 pm
Hi @ throw_away

I'm not released any game yet but as you say Torque3D and the old "Torque" having a bunch of game releases with different genres types, Tower Defense, RPG, MMO, FPS, RTS, MMORPG, MOBA, educational games, even religion games... multiplayer, singleplayer, 2D, 2.5D and 3D... and have some books published. Torque3D or old Torque books

Don't worry about torquescript, at first I did, but is not difficult at all, is a scripting language...
i still don't get the whole client/server thing everywhere....For a game that is supposed to be single-player only, it's quite confusing what needs to be server-side and what needs to be client-side
ok I understand how you feel, but again don't worry about this unless you have future plans to add multiplayer. At your first singleplayer game do the scripting as you wish and learn from this. (Torque 3D will manage with the local network side for you)

A lot of people said what of the strongest part of the engine is the network, for that reason a lot of released games are multiplayer, like MMORPG with persistent world. I know a FPS MMORPG games (http://lifeisfeudal.com/) and a Car style MMORPG (http://www.dark-wind.com/) but it should not necessarily be like this, see this games http://tiltedmill.com/hinterland/ and http://www.dimensionu.com/dimu/home/Games_Dimu.aspx and http://www.towerwarsgame.com/ and http://www.deadstate.doublebearproductions.com/media/ see this link: Games Made With Torque 3D
how difficult is it to make a game with torque that is NOT a multiplayer first person shooter?
For a sigleplayer only... very easy.
That is NOT a FPS.... What kind of game what you want? You need to know TorqueScript.... The dificulty is about you programing experience, how quickly you can learn, etc, etc... This is true for anything. So if you know how to change thing will be easy.

Of course here the game engine can help you, if the game engine is like sprints RTS well maybe you can do a RPG but you need to change some parts of the engine or maybe the whole thing... The point is... the old Torque was development with FPS multiplayer in mind, thougth the year was evolved one way or another. So a lot of stuff are focused FPS multiplayer. But again a lot of people published a lot of diferrent games.
Would you suggest sticking with torque and learning more about it and it's internals until it all "clicks" or would urho be a better option since it seems easier to get started with?
Torque3D have a good roadmap ahead and is very active, almost daily are receiving updates, upgrades or fixes. (Some are on unofficial repos to then move to the official torque3d repo)

So in anything we can help, let us know.

Cheers
John
Last edited by Johxz on Tue Oct 04, 2016 9:23 pm, edited 1 time in total.
throw_away
Posts: 8
Joined: Tue Oct 04, 2016 4:27 pm
by throw_away » Tue Oct 04, 2016 8:26 pm
@
User avatar
Johxz
Thank you for the detailed response. The roadmap does indeed look promising with entity-component system and pbr in the works!
Johxz
Posts: 447
Joined: Sat Feb 07, 2015 11:37 pm
by Johxz » Tue Oct 04, 2016 11:22 pm
@Johxz Thank you for the detailed response. The roadmap does indeed look promising with entity-component system and pbr in the works!
Indeed man... Either way.... Torque3D or the old Torque have proof that you can do different types games and finish it, and that list I shared to you is incomplete, I have to update it :oops:

I give you a categorized list of game.

Games:

- Open world action game: http://www.moddb.com/games/deadlymatter (Is in development, I think is like Far Cry, I not sure)
- Turn-based/Rol: http://irontowerstudio.com/age-of-decadence-news
- Card Game: http://www.garagegames.com/products/bellatorus
- Sport combat car game: http://www.metaldrift.com/
- Strategy game: http://kingmania.300ad.com
- Melee Duel Arena: http://www.mode7games.com/content/game.html
- real-time strategy/action hybrid: http://www.martianarctic.com/games/cyberwing/ (Is a MOBA or ARTS)
- RPG: http://store.steampowered.com/app/314120/
- Narrative/Religious: http://interactiveparables.com/ (about the bible)
- History: http://tiltedmill.com/mosbys_confederacy/
- A Torque3D Modification for RPG games: http://store.steampowered.com/app/329890/
- A Car game and Torque3D Modification: http://store.steampowered.com/app/284160/
- turn-based tactical sport game: http://store.steampowered.com/app/237350/
- Interactive and educational virtual reef: http://joticarroll.github.io/Torque3D/
- Sandbox game: http://store.steampowered.com/app/250340/
- MMO/VR/Exploration/Crafing game: http://steamcommunity.com/sharedfiles/f ... =446054581 (Is in development, I don't know whats going on with this game :lol: )

Some games use Torque3D before MIT other games after MIT ;) I hope it helped you decide man.

If someone tell to you that you can't do it in Torque3D they are very wrong :lol: :lol:

Image
L3zak
Posts: 5
Joined: Mon Jan 11, 2016 7:09 pm
by L3zak » Wed Oct 05, 2016 12:52 am
Hi,
I’ve spend some time testing both engines, so maybe this post will help You a bit. I won’t write long comparison, since after reading Your post I guess that You are aware of main pros and cons of both engines.
Basically both engines are different and designed to be used in different ways, I guess You will have to choose what fits You better. Urho gives You better flexibilty – You can use it as an external library and that will give You control of application from begging to the end, but will also require to write a lot of stuff from scratch. On the other hand in Torque You build upon existing template – a lot of stuff is already handled but You have to use it in right way. Let’s take level loading for example: In Torque You call one function (loadMission(%misfile) if I recall correctly) and lot of stuff is already implemented – You will see splash screen with loading status and after that You game will be set up. Additionaly You can add some logic in script (for example in function loadMissionStage2()). If You want the same result in Urho You will have to write everything from beggining – set splash screen, load file with mission, add it to scene tree, set input, set camera, set viewport, set player, run initialization logic. The same goes for other things – in Torque you have whole game framework build from core scripts (You can of course modify them), in Urho You’ll have to write this framework. Some people may prefer first approach, other may prefer latter. You want to use other view then first person? Torque has several build-in camera modes You can use them, but there may be need to rewrite some parts, becouse You want it to work a bit different – in Urho You’ll have to write it from scratch so it I will work how You want it to work from the very beggining. Player, vehicle, weapon, item? Every case is the same.
Also note that upcoming changes in version 4.0 (entity-component system, new base template) will propably improve flexibility in Torque.
Other users may throw stones now ;) but when I had Your dilemma, I've decided to go with with Urho... personally I prefer flexibility (but I don't mind Torque's approach) but the main reason was performance - in this field Urho is just much better, I'm using not so new computer and in my project I don't need Torque's great environment tools. Of course I'm still keeping my eye on this engine, because I really like it.
throw_away
Posts: 8
Joined: Tue Oct 04, 2016 4:27 pm
by throw_away » Wed Oct 05, 2016 10:18 am
@ Johxz Thanks, you've convinced me to spend more time with torque and experiment some more. Let's see where that gets me! :D

@ L3zak You're completely correct about both engines and I too prefer flexibility and doing things my own weird way(I prefer emacs over an ide and only use the ide to compile :lol: ). However, since i'm going to be working alone for most of this project i wouldn't mind some handholding. In an ideal scenario i'd start from scratch using only opengl and the platform's own libraries and build my own lean and mean game that does exactly what the game requires and no more(Low level programming can be so much fun!). But after starting several projects in a lot of different frameworks/libraries/engines and abandoning them later, i now want to finish something. There are things finishing something can teach that can't be learnt even if I start a thousand new projects and never finish them. One weird thing though, i got better performance running torque on linux with basic lighting than urho even with shadows low of off. Maybe i was doing something strange :roll:

Anyways, thanks for the help everyone. Now comes the part where i have to read a billion docs, samples and source code :geek: . I hope to have something to show and post at the forums soon!
Duion
Posts: 1319
Joined: Sun Feb 08, 2015 1:51 am
 
by Duion » Wed Oct 05, 2016 10:51 am
If you want an open source game engine, there is no real alternative to Torque3D, for small games like 2D games or mobile style games there is a bunch of alternatives now, but for large scale 3D projects they cannot compete yet.
So it depends on your project, if it is a small game Godot or Urho3D may work better, since you get results faster, but in the long run you will be missing features that Torque3D provides by default.
throw_away
Posts: 8
Joined: Tue Oct 04, 2016 4:27 pm
by throw_away » Wed Oct 05, 2016 11:26 am
@ Duion Yeah you're right. Godot seems pretty good for 2D and the tools seem nice as well but their 3D is still a work in progress. Urho seems well designed but so far no shipped titles and a lack of good tools. It seems like a good starting point for building an engine on top of though. There are a lot of other choices as well but most are either unmaintained or just toys that their author lost interest in. Idtech 4 is interesting but if you haven't fiddled with id engines in the past then it has a significant upfront cost of getting to know the code-base and tools etc since there are almost no tutorials/documentation available and you basically have to look at doom 3's assets/code and reverse engineer everything. All that leaves Torque3d as the only real option for me, it has a mature toolset, a number of finished games to it's name which proves it's the real deal and a small but helpful community.

There are a lot of "engine X vs Y" threads on the interwebs meaning a lot of beginners seem to have analysis paralysis. Maybe this will help someone else in the future as well!
I'll be sticking with torque for now though :D
LukasPJ
Site Admin
Posts: 409
Joined: Tue Feb 03, 2015 7:25 pm
 
by LukasPJ » Wed Oct 05, 2016 2:33 pm
If you're sticking with Torque, remember that we have a small but helpful Wiki as well:
http://wiki.torque3d.org/

It's not that widely used, but there are some useful information in there. Especially for getting started.

I wrote this beginners guide which has received a lot of positive comments:
http://wiki.torque3d.org/scripter:coincollection

Furthermore, you seem like me, the kind of guy who hates when there is too much stuff in your project, and you are not completely sure what does what (I can't stand auto-generated code and templates :P )
If so, take a look at this repository:
https://github.com/crabmusket/t3d-bones ... /tutorials
Which has a lot of very useful tutorials for starting from scratch.
throw_away
Posts: 8
Joined: Tue Oct 04, 2016 4:27 pm
by throw_away » Wed Oct 05, 2016 2:46 pm
@ LukasPJ Yeah as a matter of fact, i do hate auto-generated code and a lot of stuff that i don't understand in my project tree :lol: I've looked at all these resources and others before(i've had my eye on torque ever since it went opensource :P ) and t3d-bones framework has indeed proved very valuable! Thanks
23 posts Page 1 of 3

Who is online

Users browsing this forum: No registered users and 5 guests