It looks like you're new here. If you want to get involved, click one of these buttons!
Originally posted by Gintoh Lets just say I was making a sandbox scifi mmo (tabula rasa esque combat but SWG esque sandbox/ city building etc. mechanics) and this game had a 10 mil budget. How much time would I save using Hero engine save vs building an engine?
there are a lot of factors that go into how much time would be saved (how numerous / complex the features you want are). Assuming Hero did everything you wanted, though, the amount of time saved would be in years. And the amount of money saved would be in millions.
When building a new engine, most companies make a smaller / proof of concept game to recoup some of those costs.
Ask the SWTOR development team about the HERO engine..
Most of their budget and time was lost on the Hero engine and trying to moddify it to do what they wanted, and in the end it just could not do that..
Any top of the line MMORPG should have an inhouse engine... Using general engines ingenerall ended up with failing projects... SWTOR, Vanguard the most wellknown examples of this...
Best MMO experiences : EQ(PvE), DAoC(PvP), WoW(total package) LOTRO (worldfeel) GW2 (Artstyle and animations and worlddesign) SWTOR (Story immersion) TSW (story) ESO (character advancement)
Originally posted by Lord.Bachus Ask the SWTOR development team about the HERO engine.. Most of their budget and time was lost on the Hero engine and trying to moddify it to do what they wanted, and in the end it just could not do that.. Any top of the line MMORPG should have an inhouse engine... Using general engines ingenerall ended up with failing projects... SWTOR, Vanguard the most wellknown examples of this...
People have already said numerous times in this thread. If you had bothered to read, you would know that the version of Hero Bioware bought was an Alpha version. They had to modify the shit out of it because it wasn't even finished.
Modern Hero engine is a completely different story and what SWTOR uses doesn't even resemble Hero engine as it was then or now.
You have $10M and you cannot do such a crucial piece of research and resort to gathering amateur opinions on internet forums?
This is the kind of issues that people do technical research, and write 10 page reports for.
Originally posted by Gintoh Well I can but I just wanted to start a discussion because it would be fun.
well .. if fun and amateurish opinions are what you are looking for ... then here is my two cents.
I say you will save 6 months 3 days 2 hours and 16 minutes.
Originally posted by nariusseldon Originally posted by Gintoh Well I can but I just wanted to start a discussion because it would be fun.
I think Narius grossly over estimated. You'll save at most 3 months, 12 days and 9 hours.
There isn't a "right" or "wrong" way to play, if you want to use a screwdriver to put nails into wood, have at it, simply don't complain when the guy next to you with the hammer is doing it much better and easier. - Allein"Graphics are often supplied by Engines that (some) MMORPG's are built in" - Spuffyre
Originally posted by Distopia Originally posted by Lobotomist Hero engine is bane of MMO gaming. All MMOs that used it crushed and burned.
All, lol..? You folks really need to read up on this stuff before stating things you obviously know very little about.
While I agree using the two games released on the hero Engine ,with one using a heavily modified version and the other developed by a company that is very hit and miss with everything they do,to villify and condemn the engine itself is foolish...until there is a killer game made with this engine or several I don't see the point in praising it and singing it's THEROETICAL virtues either
At this point in time both the current Unreal and Unity engines are more proven the the Hero Engine,that doesn't mean they are better or worse just that the hero engine has not given any evidence of being better and two dubious examples o being worse.
CSE decided to make their own engine for Camelot Unchained. Luckily they have a very gifted developer heading he engine build with experience. I can only imagine the knowledge heading into building a game engine is most important. The kickstarter was an early one and truely for kicking off the project. Due to this we are able to see everything progressing including the engine.
So far I would say they easily have added at least 6 months to a year because of developing an entirely new engine. It also was noted they had developed engines in the past using a different language and wanted to head in a different direction which I have no idea if this saves time in the long run or not. Other elements of the game are progressing at the same time but man hours are diverted for engine development no doubt. Parallel this with net coding, concept art, early model development and mock ups using different engines/apps and the first year is mostly prep work.
CSE hopes this actually saves them time and money in the long run over leasing an engine that may or may not meet their performance requirements and likely the is the most important question. A company leasing/buying an engine would already be building on top of that engine by now but then a great deal of trust must go into that engine to ensure it doesn't limit your ultimate goals. A good example of this is Pathfinder. They decided to use the Unity engine and have therefore already released vids on world building. Camelot Unchained has only been through net coding and base engine testing not yet even using actual CU world elements. Cost, development time and final performance weigh heavily into factoring which direction to take.
As for the Hero engine my opinion of it is it's developed from the ground up to make the themepark games I hate so will never support another Hero engine game. Even if used otherwise it always seem clunky and slow for my liking.
You stay sassy!
Originally posted by Loktofeit Originally posted by nariusseldon Originally posted by Gintoh Well I can but I just wanted to start a discussion because it would be fun.
May be averaging the answer will be more accurate. The book Wisdom of the Crowd says so.
I would use this before i even thought about that engine.
Originally posted by Gintoh Well, let's just assum it did have everything, what is a long time? 6 months? A Year? 2 Years?
If the Hero Engine--or any other off-the-shelf game engine--has everything that you want, then your game is a generic clone and should be canceled immediately. Don't waste your time looking for an engine that has everything that you want.
That's not to say that you should never license a game engine. If a game engine does much of what you want, does it the way you want it, and does it well, and you have the capability to readily modify or add to it to get the rest of what you want, licensing the game engine may save you a lot of time and effort. But it's important to recognize that you will have to modify the game engine. If you don't have that capability, don't start the project.
And to modify the game engine, you're going to need access to the full source code. Game engines might say that you can license it for a few hundred dollars or some such without the source code, but that's basically a trial version. If you want to actually launch a game, you need the source code, and you'd better find out ahead of time exactly what that is going to cost so that you don't get shocked by an unexpected six figure price tag later.
Choosing between game engines is not a question of looking for the one that does everything you want. Rather, it's a question of which game engine would be the easiest to modify into what you want. If you're going far off the beaten path, building your own from scratch is likely to be the easiest there. Building your own means that when you inevitably need to make modifications, you already have someone on the project who knows exactly what the engine does and why it was set up to do things that way--because he coded it himself. If you're less ambitious on the technical side of things, licensing a game engine may make more sense.
Originally posted by Ridelynn Well, as far as time saved... I guess it depends. They've spent what, 8-9 years developing Hero to the point it's at now? But that's an engine designed to be packaged and sold as an engine, if you jsut developed the features you needed, and didn't bother heavily with the support and documentation that a commercial product needs to ship with... maybe half that time? The original id tech engine was developed over the course of several decades, starting back in the early 1990's. That's powered everything from Wolfenstien 3D to Rage. CryEngine and Unreal Engine are in similar boats, developed and refined over a number of years, and iteratively released as they get access to new and more powerful hardware. So.. if you just wanted to move pixels on a screen, you could write a 3D engine that uses basic textures in less than a year, but it will look like something out of 1991. If your wanting fully features shader engines and network code and the entire works, your looking at years of development. And even with a fully mature engine, your still looking at 18 months minimum just to plug in your art/music assets and get the AI and such working how you want it... and the more you deviate from the building blocks of whatever engine you pick, tack on just that much more time. 10M is a lot of money, but for an MMO that needs a back end capable of handling thousands of users, a marketing & distribution network capable of selling to thousands of gamers, and multimedia assets to make it look and sound interesting, your out a lot of money already before you even program one single line of code.
You touched on this, but there's an enormous difference between making an engine for your particular game and making an engine that is intended to be licensed to others. In the former, you can often say, I don't need this capability for my particular game, so I'm not going to implement it. In the latter, you often have to say, a lot of games will want this capability even if most won't, so I have to implement it. Having a lot of things that you don't need to implement can easily simplify coding the things that you do need, too.
Also, if a game engine is in development over the course of decades, you're looking at an awful lot of deprecated code. Going from purely 2D methods to the early 3D fixed function pipeline to early 3D programmable shaders to adding geometry shaders and the tessellation stages changes a ton of stuff every step of the way. If you build your own from scratch, you don't have any legacy bloat, nor any bias toward doing things a particular way because it made sense 5 or 10 years ago, even if it doesn't make sense today.
Making a game engine can take a lot of time in the sense of the gap between the date started and the date finished. But it doesn't necessarily take a ton of man hours. A game with a large enough budget easily have 100 artists creating artwork simultaneously. But you can't have 100 people coding the low level stuff in your game engine without constantly tripping over each other. Rather, if you're going to create your own engine, you have only a handful of programmers to start, then hire a bunch more people to join the project later on.
Originally posted by Smikis it would save you a lot of time, but your game would ultimately fail, so whats the point , not sure why you are asking those question in first place..
You kick puppies don't you?
Originally posted by JC-Smith Originally posted by Phry
Bioware had to do 'massive' amounts of work on the engine itself just to do what it does with it, and while this is using an earlier version of the engine, even after all the work they did, it still doesnt scale that well, so much so that open world PVP is a huge problem, and we're not talking even about 'hundreds' of players, but just a few 'tens' of players in an area.
Couple notes here. First Bioware bought a version of the engine that wasn't ready to be sold. It was being developed for their in-house game (Hero's Journey) and Bioware approached them about licensing it. They knew they were getting an incomplete engine. And while Hero Engine continued to be developed for years from that point, Bioware didn't receive any of the updates as they had forked out on their own and stopped taking updates, as is common practice.
That having been said, the issues that TOR had with Open World PVP have little to do with Hero Engine. They have to do with Bioware's design decisions. Bioware put a huge amount of customization into TOR and used poly counts that were much higher than most other MMORPGs. There was nothing wrong with those design decisions. Their target audience was solo and single groups of players, small raids. Given their target audience their decision made sense. Players want customization. They want the models to look good. It wasn't a game that was really designed with Ilum in mind as a focal point.
Of course all that customization does come at a price. There are a lot of state changes, and a lot of textures being used. If their focus was open world PvP they probably would have given less clothing options and used lower polygon models. Keep in mind in an MMO your dealing with multiple parts to make up your appearance, and those parts are generally each using their own texture. This makes instanced rendering of characters difficult, and instanced rendering is much more efficient. This problem grows multipliciously, so the more customization you have the less likely it is that players will be sharing the same models and textures, and the less efficient it gets for each additional character on screen.
As a developer you can't really have both, it's a compromise. The more customization and the higher resolution textures you use, the less efficient you will be rendering, but the better you'll look. If you go the opposite direction you will render much more efficiently, but players may complain that your game looks dated. Or you can try for some place in between. That's a balancing act that you the developer needs to make. I think TOR's decisions made sense for their game. They catered to what 95% of their players would be actively doing most of the time (questing, leveling, small groups). But those decisions also are a big part of the reason that the game chugs in a busy Ilum. Yet Ilum is generally what gets brought up when people criticize Hero Engine. It's just not a fair evaluation.
Last but not least. It should be noted that while Hero Engine has its own renderer, that's just a small part of what it is. It's the other features that make it an MMO engine and not a rendering engine. If like in the OPs situational post he had $10M to work with, he could if he so desired replace Hero Engine's renderer with Cry Engine, Unreal Engine, or whatever other tech they desired. It would be a bit of work, but the point is that rendering is just one of the things that the engine does. When people criticize TOR's performance in Ilum it's general the rendering that they are referring to, and Hero Engine 2 has a vastly improved renderer from when TOR was released. Not to mention TOR did a number of their own changes to the renderer.
While I mostly agree with you:
1) No matter what game engine you use, and whether it's an early version or not, you're going to have to make changes. Once you make the changes that you need, the final product is on you the developer, for better or for worse. If you can't make the changes you need, that's your fault, not that of the game engine. While some game engines are better or worse at various things, it's not that easy to tell that from completed games using them. I don't know this, but I suspect that EA greatly underestimated the difficulty of modifying the Hero Engine to do what they needed.
2) Higher resolution textures hit video memory capacity hard, but aren't otherwise a major performance hit. And it's completely trivial to go from higher resolution textures to lower resolution ones via mipmapping. But yeah, higher polygon counts (I prefer to call it vertex counts for technical reasons, but it's very strongly correlated with polygon counts) bring a big GPU performance hit. And needing more textures and more vertex arrays and more uniforms and more draw calls to draw a character brings a huge performance hit no matter what hardware you have. It's easy to draw 1000 characters if every character is a simple 2D sprite. As you said (and as with nearly everything else in life), there are trade-offs.
Originally posted by Xiaoki Originally posted by Jean-Luc_Picard Don't use the "hero" Engine. "Unity" is MUCH, MUCH better.
Maybe if the MMO was on Android.
Given the capability to modify anything you want, however you want, you could turn any game engine into what you want. After all, what is making your own game engine if not turning an empty game engine into what you want?
Of course, at some point, the changes you need become so great as to make licensing the engine a bad idea.
Originally posted by nariusseldon Originally posted by Loktofeit Originally posted by nariusseldon Originally posted by Gintoh Well I can but I just wanted to start a discussion because it would be fun.
Averaging guesses will often reduce the standard deviation of your error by the root-mean-square/arithmetic-mean inequality, but sometimes everyone or nearly everyone is off in the same direction.
Originally posted by Quizzical Originally posted by Xiaoki Originally posted by Jean-Luc_Picard Don't use the "hero" Engine. "Unity" is MUCH, MUCH better.
Maybe if the MMO was on Android.
i had thought, perhaps wrongly, that a number of mmo's had used the core of a licensed engine and then modified it to fit their needs more closely and that this was typical of the industry.
"There are at least two kinds of games.One could be called finite, the other infinite.A finite game is played for the purpose of winning,an infinite game for the purpose of continuing play."Finite and Infinite Games, James Carse
Look at it this way...
Funcom and their DreamWorld engine.
A pretty large studio that attempted to make their own engine and succeeded in making it, still failed miserably when it came to meeting expectations of both players and investors.
They had the money, the manpower and the experience. They even had the time, the engine was developed for about 6-7 years before TSW used the final version. AoC used the engine after less development time but the engine was very much in beta back then and contributed to AoCs lack of success.
Originally posted by nethaniah Seriously Farmville? Yeah I think it's great. In a World where half our population is dying of hunger the more fortunate half is spending their time harvesting food that doesn't exist.
Originally posted by JC-Smith
Of course all that customization does come at a price. There are a lot of state changes, and a lot of textures being used. If their focus was open world PvP they probably would have given less clothing options and used lower polygon models.
Their problems were not clientside mostly but their netcode just sucked (or sucks didn't check in a while). Only if their serverside is vastly improved over the version they sold to BioWare the "new" HeroEngine will fare better.
This was only one of many strange decisions on BioWares side. Getting a proven engine with actually scalable and working server implementations where you have to develop some tooling yourself is always cheaper than trying to beat an unwieldy engine into submission which you just chose because of some shiny tools. And given the money they burned on SWTOR they could have employed a lot of top notch serverside and tool developers. But they chose to instead make voiceovers for even the most generic quests...
Originally posted by aspekx Originally posted by Quizzical Originally posted by Xiaoki Originally posted by Jean-Luc_Picard Don't use the "hero" Engine. "Unity" is MUCH, MUCH better.
Maybe if the MMO was on Android.
Most AAA MMORPGs, and also most indie MMORPGs that I've liked, make their own game engine for their particular game, or perhaps make a single in-house engine that will be mostly shared between a few games. That gives you the maximum amount of flexibility to do whatever you want.
For example, the top four rated games on this site as of this posting made their own engine, as do six of the top seven most visited today, with SWTOR as the lone exception--and that's a game often derided as having a bad engine.
Smaller, less ambitious games (e.g., single-player indie games for Android or iOS) are far more likely to license a game engine than to make their own.
Originally posted by Quizzical Originally posted by aspekx Originally posted by Quizzical Originally posted by Xiaoki Originally posted by Jean-Luc_Picard Don't use the "hero" Engine. "Unity" is MUCH, MUCH better.
Maybe if the MMO was on Android.
cool, thanks for the info.
If you don't know how to answer your question, you're not ready to make an MMO.
Now aside from the truthful trolling. It's going to depend.
Getting something "worth showing" is going to be pretty darn fast using someone else's engine. IE getting to the point you can start showing news sites, investors, motivating your team*, and similar.
When you're at the point where you start to "do your own thing" it starts to become a lot more interesting with someone else's engine. This is the point where you start having to work around someone else's work, I've seen supply nodes hacked into engines that don't support like this: New enemy type, 1 health, immune to all damage except "pick", Some hacking so the pick is allowed by all classes, switches automaticaly when harvesting, and doesn't show the stats of a weapon. That's a really really really simple work around, if you have something more complex like needing the system's clock to run faster/slower for a new effect/efficency You're looking at months of modifications, training, and testing(by hand going over every last piece of code in the engine and with understanding).
It really depends sometimes adopting an engine for a project can lead to more time and frustration being required.
*probably the most important factor in an indy team. And if it isn't it's in the top 5 at least.
Practice doesn't make perfect, practice makes permanent.
"At one point technology meant making tech that could get to the moon, now it means making tech that could get you a taxi."
Originally posted by anemo This is the point where you start having to work around someone else's work, I've seen supply nodes hacked into engines that don't support like this: New enemy type, 1 health, immune to all damage except "pick", Some hacking so the pick is allowed by all classes, switches automaticaly when harvesting, and doesn't show the stats of a weapon.
This presumably isn't news to you, but that's exactly the sort of kludge that you want to avoid if you can. That leads to counterintuitive, unmaintainable code that could easily cause really crazy bugs years down the road when it doesn't play nicely with something that someone else changes. Building your own game engine means that you don't have to resort to hacks like that.