Xenko game engine developer buys Geomerics, developer of Enlighten

That's true. But it is best I can offer for some quick research. But I also see games on sites like GameJolt... the number that have the Unity startup screen have always been a tiny percentage. And LudumDare also you'll find games made in a large number of different engines and apis. Mainly I just know there are a huge number of game devs out there using other things because I visit those other communities in my quest for the ultimate tools. lol

 

Googling around I can find numbers for Unity anywhere from 34% up to 62%. Nobody seems to have done a statistically valid global study. But there are plenty of studies done on market segments. All of them seem to show that Unity is fairly dominant over any of its competitors.

 

Enlighten is very bad for my point of view, so very happy if this guy left Unity , and good luck for Xenko

 

I find that very hard to believe. Not that you googled and found mention of those stats but the stats themselves. Maybe if the stats are for % of mobile games made and they only looked at the top 5 to 10 engines then yeah. Because I have seen so many people use Unity to throw out just anything on app stores. At one time I actually checked out games on mobile and came across Unity "games" that were things like a model rotating on the screen or multiple games that are the Rollaball tutorial.

 

Back in October 2016 Evans Data estimated the number of mobile developers (just about everything including game development is part of their survey) at 12 million.

http://www.eweek.com/development/mobile-developer-ranks-reach-12m-to-top-14m-by-2020
https://evansdata.com/reports/viewRelease.php?reportID=6

By comparison in the same month of the same year Unity had 5.5 million registered developers.

https://web.archive.org/web/20161026023037/https://unity3d.com/public-relations

According to their chart the number of developers should be at least six million now but they no longer mention that statistic on the site as it isn't very useful to know since not everyone does it professionally and there are far better statistics.

 

Sometimes it's good to have a look at things from a different perspective. The following post discusses why Unity seems to be the most popular engine, from the perspective of Unreal users:
https://forums.unrealengine.com/showthread.php?98600-Why-is-Unity-the-most-popular-engine

 

Why? Other software segments see similar massive skews towards one product line. Windows dominates the OS market, Photoshop dominates in the image editing sector, Steam dominates in digital distribution. The list goes on and on.

Software tends to lend itself well to natural monopolies. Its no surprise that this is happening in the game dev world as well.

 

Thanks for that. No doubt a huge factor is mobile. I can't quite understand why now though when there are so many engines and apis for creating mobile games quickly including true drag n drop tools. But I guess like some of them said people just got used to it so stick with it. Also interesting the part about Unity getting into the schools for training. That probably hooked a lot of youngsters.

 

I can definitely believe it for mobile but not game dev in general there is just no way. lol

 

Have a look through your own games library. Open up the source files. I bet you will find far more Unity games there then you thought. For me somewhere around half of the games I play regularly on PC are Unity ones.

Whichever way I approach this problem, I come up with the same solution. Unity is big. Unity pretty much dominates the market. Noone else comes close.

 

Thinking about and looking at the numbers again I think the Twitter Followers is not a bad indicator of usage. It seems to be pretty accurate looking at Unity, UE, GMS, AppGameKit, Xenko etc. I'd say the Followers are pretty much accurate as far as usage goes comparatively speaking.

It could be very true that Unity has maybe 50% or more of all mobile game dev. That wouldn't surprise me. So things are subjective of course. UE probably does have a huge number of developers too but probably a much higher percentage of them are making desktop or console games. And some of the other engines a much higher percentage would be making web games and so on.

It might actually go like this...

Majority of all mobile game development (particularly 3D) is done in Unity.
Majority of 3D desktop and console game development is done in UE.
Large number of 2D mobile, desktop and web game development is done in Gamemaker.
Then it gets further split out across all of those other engines and apis... CryEngine probably most devs are targeting desktop and console 3D.

It does seem like mobile game dev is super popular. Has to be an absolutely massive market because Unity seems to have most of it yet there are companies with game engines that only make monile games and they are surviving and growing.

 

I'll do that tomorrow. But I think another difference is I am often downloading games, buying games at other places such as itch.io etc. And I see games made in various engines and apis as a result. I will do that though. Tomorrow, I'll check the games I got from itch and Steam and so forth.

Still there are definitely a huge number of people doing game dev in things besides Unity. All you need to do is visit these other game engines and game library sites and check the showcases and forums.

Bringing this back around on topic.... even Twitter highlights the low usage of Xenko. It is basically only half the size of the App Game Kit community and that is quite small compared to Gamemaker which is small compared to here.

 

Unity is easy to use and streight forward compared to many other 3D engines for people who have been able to learn how to use it lol

Multi platform , mecanim, navmesh options, lighting options, thousand of plugins (from FSM , visual scripting , behaviour trees etc ...), you can go to production very fast.
And it's very fast to learn with thousand tutorials on the net about any gameplay or any features unlike CryEngine or Xenko.

Anyway, Lumberyard perhaps will become a great solution , and having some more appearing like Xenko would mean more options for game developpers with their own specific stuff ( each solution having it's own advantages and disadvantages ).

You can't compare Xenko a modern 3D engine and editor (still in wip) with very small scope and very limited engines like app kit or gamemaker lol

 

instead of twitter, maybe overall forum posts (or if possible forum posts in the last 2 years) would be another interesting stat....but again, different engines for different purpose....possibly the best comparison would be an agreed list of 100-1000 benchmark games (i.e. commercial / non commercial, big studios included / excluded) across platforms and then see with which engine they were made...hard work to figure out. While it might be interesting, it does not really mean much. Each project and individual combination will have their own best solution...for me it is the evil I know.

 

Someone could do the legwork on steams best seller list from last year. Without doing any research I identified at least 11 titles that I know are Unity. There will be plenty more I missed.

 

I was curious, so I went to Metacritic, checked their quarterly report The 25 best PC games (April 2017) and tried to find what engines these games use. 9 of 25 games from this list were made with Unity. (updated numbers with Ryiah's info)



EDIT: I've also created an image with screenshots of all these games: http://console-dev.de/tmp/metacritic_best_25_pc_games_april_2017_screenshots.jpg

 

Nine. The Elder Scrolls: Legends is Unity too.

https://venturebeat.com/2017/03/08/elder-scrolls-legends-leaves-open-beta-and-launches-on-pc/
https://www.reddit.com/r/elderscrol...game_for_mac/d6meaan/?st=j482n2bc&sh=800692d5

That's 36% which is very close to the number Unity mentions on their public relations page.

 

If you separate AAA content projects from small scale games from , you won't find Unity games.
It's like mixing all possible 2D and mobile games that are published side by side with big blockbusters games, you'll have lot of small games and less AAA indeed.

 

How are we defining AAA games here? Because Steams top grossing games certainly had plenty of Unity titles.

 

Great, I updated my post to reflect this!

I've the same impression.

 

Just got activity report from xenko forums. haven't been there for a month or two now.

7 new topics
24 new posts
5 new users

Dead community.

 

hmmm...someone is trying to influence the forum stats "
买新加坡大学学士学位证、成绩单QQ微信963146376买美国大学学士学位证、成绩单、买澳洲大学学士 学位证、成绩单、买英国大学学士学位证、成绩单、买加拿大大学学士学位证、成绩单、买法国大学学士学位证、"

 

It would be nice if someone from Unity could pop in and let us know if this means anything for Unity.

I'm not looking forward to yet another transition to a different lightmapper (as much as I despise Enlighten)

(for the record, soon after Beast was bought by Autodesk, Beast in Unity stopped getting updates and then was eventually dropped)

 

What other lightmappers are there that aren't owned by a company making a game engine?

 

Very big team and budget, sometimes fully custom 3D Engine and tools.

 

It has less to do with the asset store and more with not spending 5 years writing your own tech.
Main strengths of unity is multiplatform and mobile support.

It would be great not to start another "AAA" thread again.

 

nice definition brah

 

Personally, I found their engine to be far advanced of Unity in some ways, it's just not as far along and progress is a little slower than I'd like. I get the impression they have fewer people working on the engine. You'd have to do a lot of the shaders and such yourself, instead of rely on an asset store.

But if it weren't for the asset store, you'd be doing the same in Unity. Want an advanced terrain? Write your own, or use the asset store.

I get the impression with Xenko that if they had the army of developers that Unity has, they'd be a stronger competitor against unreal, and Unity wouldn't even be a target.

 

Good question...

...

Probably none?

I guess that's why the progressive lightmapper is probably the future. Although even that isn't a complete Unity creation since it relies on imgtec tech.

 

That's interesting for sure. I think it would be better to take a sampling of all games released in the last 12 or 18 months. For desktop. For console. For mobile. For web. It would take a lot of work but whatever games are identified could then be researched further to find what was used to develop them. I think that would give a clearer picture of game development engines and apis being used.

Who knows maybe Xenko has powered a few more games than we think.

 

I guess the simplest approach is just looking at the different engines websites and see their showcases. This wont be representative, but possible a good representation of what the engine developers think is good marketing for their tool. This will likely also give some indication as to the strength of specific engines...

@GarBenjamin do me a favor and develop a game instead of going on a massive game engine research mission

 

lol

 

Lol Dan do me a favor and make a game instead of telling me to make a game.

Two of my current projects are games and I actually worked on the 3D shmup this morning. Need to do a YT vid for this update later today.

I'm just very relaxed about it all. When I work I focus and work and after those 20 to 45 minutes are done I forget about it again until next time I get in a game dev mood. This morning I got in 30 minutes before work. It's starting to take shape bit by bit.

 

This one-man-army-engine too : http://www.s2powered.com/copiasito/

 

Wow. Blasted by horrible music upon opening the page... found the winner of the worst website award for an engine.

 

To be honest, it should be probably possible to bake level in blender, or readopt some of the quake 3 era tools. There seems to be some sort of lightmapping library on github too.

 

Yes, but there is no established workflow for doing this and Unity doesn't facilitate it at all.

It's doable, people have done it, but a good built in lightmapping solution will always be much more convenient.

 

Just found a couple more news articles both mentionining that Enlighten will continue being licensed out to third parties.

http://www.gamesindustry.biz/articl...studio-acquires-enlighten-tech-from-geomerics
https://mynintendonews.com/2017/06/...-will-be-compatible-with-the-nintendo-switch/

Research Tycoon. A quick search shows no results. Have at it!

 

I beleive in one of the "give us Beast back!" threads, a unity developer was talking about possibility of integrating third party lightmapper into the project.

I have hard time finding the original thread at the moment, though...

 

Arm was bought by a Japanese company not so long ago, so it becomes easier for deals like this to take place. I am not worried. Enlighten is not an essential tech, and is becoming less relevant as GPU power grows.

In any case one can bake extra data out to probes and use those to propagate/relight the scene. I have had some successes with quick tests.

No concern here.

 

I think there are a lot of them out there done by one person. And some by small teams. And many that are open source. Every time I try to find an complete list they never are. Another I've seen mentioned for games lately is something called Superpowers game engine. Seriously that is the name.

There are hundreds. Maybe even thousands of them out there. One reason I check the Twitter accounts is because often game devs will tag their post with the engine they used and even this Superpowers seems to have people actively creating games.

If all of these people got together... probably nothing would be done due to disagreements... I mean if they all could focus on one like Xenko they could make something awesome.

It's actually impressive the number of people out there making these game engines, languages and libraries from scratch or branching existing open source projects.

At any rate I see no reason to use Xenko over anything else at this time. I think I finally found the best one for me for most 2D and 3D game projects in AGK2 and probably N6 for 2D pseudo-3D raycasting games. Doesn't look like Xenko is close to either.

 

And I believe I also replied in that thread that, yes you can do it, but the workflow will be a nightmare

Plus, a lot of the features are interconnected. If you don't use builtin lightmapping, you also can't use lightprobes, you'd have to write your own solution for that. And most probably you'll need your own custom shaders. And replacing all those built in features is a lot of work, which then need to be maintained a lot because you'll need to dig deeper into editor apis and those break a lot more often than most people think.

Plus, the more Unity conveniences you throw away and write yourself, the more you start wondering why you're using Unity in the first place.

 

Yet another platform built around HTML5 and packaged using Electron just like Visual Studio Code. Back when I first heard about Electron it occurred to me that it would be an ideal way to support multiple platforms with an app or game framework without having to deal with the headache of different platform APIs.

 

Might be a good idea to dig that thread up. I forgot where it is.

Multi-platform support, scene editor, existing apis for large number of tasks. Even if you throw away EVERYTHING, well, multiplatform support will still remain.

 

Most game engines / dev platforms have at least some degree of multi platform support these days. Heck BlitzMax had it long ago. Unity may still be the king of it though but how many of those platforms will you actually target?

Steam gamers seem to hate mobile game ports. But easy multi mobile (which means Android and iPhone okay probably Windows too) platform build support is good if making mobile games. Windows, MAC and Linux support is excellent if making desktop games. Xbox and PS support is great if making console games. Web (HTML5) is great for everything if only to release tiny playable demos for feedback.

I think in actual practice the majority of devs probably only need support for mobile + web or desktop + web. So I guess Windows, MAC, Linux, Android, iPhone, Windows (mobile) and web would cover the majority. Of course there are devs who want to release on Xbone or PS4 and handheld consoles but I am talking about what the majority are actually doing.

 

It is not a good thing to end up in a situation where you need one more platform, but have to spend few months bolting on support for it.

The reason to use unity is that you don't need to integrate physx yourself, don't need to write your own animation controllers, don't need to write your own model loaders, don't need to implement your own scene graph, don't need to make editor and tools from scratch. That's several years of work.

Steam gamers that hate mobile games are non-factor, because they wouldn't buy such game in the first place, meaning they're not a target audience.

Now, it is entirely possible to keep dropping unity features to the point where you can drop the whole engine, but getting to that point takes time and a lot of work.

The main issue is that support for mobile and other exotic platforms is guaranteed. since the beginning of the project. With different kind of tech, it is often "It'll probably work". Trying to set up libsdl application for webgl is probably going to be a lot of "fun". And I think isntead of dealing with this kind of "fun" but pointless technical task, it would be better to concentrate on game development instead.

And that's the reason to use unity. Someone already did several years of work for you, and you can just stick with strictly game-dev realted task at hand.

 

I often speak with slight hyperbole to drive my points home. Yes, of course, there are still plenty of things that Unity does for you.

But I've found myself working around Unity's systems and features far more often than I'd like. I mean, obviously more generalised features can't be tailored exactly to my needs and they'll never be as fast as a solution that does exactly what I want it to do and nothing more. But still, we say to each other "that feature is too slow/buggy/does-not-cover-our-needs, let's roll our own" far more often than I'd like.

 

True, but such is life i suppose. The thing about unity is that you had that buggy feature you replaced available, and it allowed you to get the prototype to the point where the feature had to be replaced.

Basically, imagine you bought a car, and then started slowly customizing it to better suit your needs. In the end you may end up in situation where there's nothing left of the original car.

However, if in the beginning you had only a pile of parts and not a car, then you wouldn't be able to get your custom car that easily.

Hopefully this makes sense.

 

If you need to release a game on an exotic platform then use Unity (even though some engines supports a truck load), simple..! If your viable platforms contain the usual (Web, Console, PC, Mobile) then you've got a lot of options.

Most engines don't require you to hook your own physics systems, as for tools well Unity is near enough as bare bones as it gets (really) which even a mid / large sized indie wouldn't be able to get it on equal grounds with UE within a reasonable time frame.. Besides the recent iteration of Enlighten the few systems available are aimed towards static (smaller) games with a variable level of quality to boot.. Not saying they aren't trying and / or making improvements but we are still many years off..

Then you start grabbing for the asset store (for a workaround) which in many cases requires a workaround for a workaround, I've had to constantly build stuff for Unity too so it's far from strictly "gamedev". I get you're talking about building your own from scratch but with so many options I'm not really seeing the point in it.

Ultimate point? Well this is where Xenko somewhat fails to me, I've had a couple of stints with it and yeah in some areas it is more advanced than Unity. Then again so is Stingray and most seem to have forgotten it exists.. Xenko trying to copy what's essentially a multi-platform game framework (which is really what Unity is) without all the bolt on's from the asset store and without the years of community support / documentation / tutorials / examples seems somewhat illogical.

It's like me saying I'm going to copy W3 but actually it will be ten times smaller and I'm replacing Gerald of "Div"ia with a duck, oh and I'll charge you full price for it too. You'd have to really dislike Unity and UE to go there.. Buying Enlighten won't change much, I get the feeling they are banking on asset store dev's porting over all their stuff because it'll be relatively(ish) straight forward to do..

It worked for Unity because at the time there was no competition, community orientation / simplicity in areas like coding made it successful.. It's kind of like the Skryim of the game engine world, Unity is the creation kit and we're the modders..

Not saying they shouldn't try, it could in years to come be an awesome replacement but for now I don't think many will ditch their Unity pipeline for Xenko. As for Enlighten, sounds like a funding element for their own engine..

 

It makes sense, but it's not completely analogous. Since parts of the car, like saaaay, the engine, are complete black boxes I can't do anything about (I realise Unity is working on opening things up) and if I want things fixed/better in them, I have to use newer versions of the engine. Which would be fine, but sometimes I find out that the new engine is not compatible with my custom parts any more so I have to tweak them again/throw them away.

But bah, whatever I get what you're saying and I'm not really disagreeing in a fundamental way, so let's put the car analogies to rest