Why Unity 5.0 is STILL a good deal

I find Unity's subscription plan idiotic, it makes no sense to me. The price is too high for what it is, and even after you give them the same amount as the full price for pro, you don't own anything. You have nothing if you stop paying. I would never sign up for that. If you want Unity, you should just pony up the $1500 and be done with it. Now is the time since you get Unity 4 and 5 included in the same deal.

 

One thing they lose if they don't subscribe is the access to the marketplace (like Unity's Asset Store).
edit: ah, Bruno beat me to it. (an hour ago )

And perhaps they feel they'll get more money in the end. More paying customers if they don't have to keep paying 19$ every month, and that they won't lose any people who'd otherwise have paid every month.

 

I wish I could say this....

 

as Unity 4 Pro license owner and looking at the "special" that came to me yesterday... @20$/mo for full access = 30 months versus $600 (no royalties) AND knowing the release cycle is about 2 years...

the cost difference is quite minimal ($10/mo difference on a 24 month cycle).. BUT with a full Pro license I have zero royalties... so being such a minor monthly difference, the risk involved is so minimal versus the potential loss in royalty fees.. I would have to go with a full license renewal.

but at current 75$/mo sub fee versus license renewal... that's a no brainer (full lic.)

but of course i'm not counting my initial investment of 1500$...

 

I would very much like to see them simply lower the cost of add-ons... both for subscription and for full purchase. $1500 for Unity Pro and 500/600 for addons would definitely make me squirm far less whenever a new whole number version is released — assuming upgrade prices continue the trend of being half of full-price. Seems reasonable to me, especially since you're required to get Unity Pro in order to have iOS Pro.

 

This has been one hell of a day. I'm still shocked.

 

100% agree with this.

 

That's nothing like piracy. If Epic didn't want to allow this behaviour, they wouldn't explicitly mention that you can cancel your subscription at any time and still use the engine.

The incentive to not cancel your subscription is that over the next year or so I'd imagine we'll see rapid updates and a lot of new content for UE4's market place. Perhaps in the future as UE4 matures we'll see a change in the subscription model that restricts usage of the engine to current subscribers but for now, I think Epic is trying to entice more indie developers into using their platform. Personally, I'm blown away. Here is a AAA engine that offers everything Unity has and more, enables mobile/web development for free, access to the source code, and is backed by a company with a proven track record of releasing quick updates/fixes (with regards to UDK.)

Edit: Just realised someone has already responded to your post. I'm going to sleep now lol...

 

Aye. I still stand by what I said, in the context of what I knew at the time. Of course it's a different matter now that I know Epic expressly allow it.

 

I love Unity. I've used it for years and love its flexibility and the companies attitude, process and people.

However, this is a major game changer. Unreal has a certain business model, and apparently they have figured out its OK to give away source code and a top game engine for virtually free to anyone that wants to try to make games. Essentially, they have 'democratized game development' in a single day. Pretty impressive and I am very curious how Unity will respond.

Just looking at the education component, our University can start teaching Unreal for $19 a month, with access to full source code. Why on earth would we not use UE4 over Unity?

To Unity's credit, UE4 looks highly derivative of Unity. But, end of the day, licensing terms matter and Unity is definitely on the defensive, IMHO.

 

Yes the interface is suspiciously similar in a lot of points

 

Although I think Unity indeed has to change their business model in response to this, Unity is certainly not objectively inferior to Unreal, as I've read some suggesting. Its still a simple to grasp engine that feels like a blank slate for your game. When I first started using Unity 4, I was able to make game mechanics faster than I'd ever been able to do in any other engine I'd ever fiddled with. I think I share this sentiment with many. Even for someone who values their own time at minimum wage, all these price points seem pretty insubstantial in the face of the man hours it takes to make a game, so isn't Unity's ease of use still a massive factor?

 

Yes it's still a massive factor, but people only see the few buck they don't spend every month...
Race to the bottom for engine has started!
Unity will evolve slower (less money), Unreal will took many years to get where unity is right now, but will have a userbase attract by the price.

 

What makes you think Unity is easier to use? It may well be, but we don't know that yet. UE4 has a completely new interface - it's not UDK.

Have you looked at the videos on the UE4 site?

 

Unity5 Pro free + 5% or i buy UE4. Waiting for the movement of unity.

 

I think Unity could keep its prices, but drop the 'up front' strategy. I don't think that's going to work any more.

Keep the prices, but back-load them instead of front-loading them. Remove the Unity Free and maintain only one branch. That will lower costs some in Unity's development. Unity can motivate the retained costs by being fixed rather than diffuse and fluid as they would be with Epic.

Release Unity Pro to all. This way we can use Unity Pro as we use Unity Free now, but if we publish we have to pay $1500 + the additional licenses.

Yes, devs might not sell enough to recoup that cost, but that's not Unity's problem. It would also be the least destructive for the Unity company plans and processes, I guess.

 

My guess is they've only set a subscription fee to create a barrier to entry, to keep the rabble out, or perhaps to have some form of business relationship in place for when the time comes to collect the royalties. A subscription fee of $20 a month is hardly likely to bring in large wads of cash for them anyway.

 

Maybe it's just me but I don't see where this is such a great deal. It's too good a deal to be true so it isn't. UDK was free with a large net profit percentage but the vast majority of indie developers don't make more than 50 K anyway. Epic aren't even interested in the 5% gross because really how would that be manageable from an administration POV if thousands of games are produced.No, Epic want the $20 subscription fee because they know quite well that thousands of games will not be produced, why? Because they never have been up to this point and they won't be after. UDK was free and had a similar toolkit only lacking the source code. So, will having the source code suddenly give the ability to produce even better games with UE than UDK? Don't think so. The reality is that UE was only ever really used by companies that would buy the source code and had the resources to put together AAA games. AAA engines are built to push pixels and all the polish it has will not translate to mobile devices where the indie market is. Plus if I was a company that had licensed the sourcecode for UE for half a million dollars or more and this happened I would be somewhat annoyed. Could it be that these companies are now scarce and Epic is looking for a new revenue stream. Unity give sme the tools to do amything I want and it is still a good deal at the cost and I just upgraded to 5.0. Last night some agent provocateurs started posting about this on the Merseyside Unity users group at first I was going to argue and defend Unity. But I decided to just ignore it because Unity doesn't need defending. If one had the resources to make AAA games and the platforms to sell them on readily accessible then I have no idea why you would be using unity anyway. But virtually no-one here has those sort of resources and the game market environment has changed so much that even attempting to make AAA games doesn't make much sense for a lot of companies now. Do you really think you can compete with Assassin's Creed and not spend millions to do it? Unity imposes a certain rigour and realism and makes things possible. To me this announcement is not a game changer at all. Pretty certain Epic aren't a charitable organisation and they're not doing anyone a favour but themselves. I await the flood of fabulous indie games made with the holy grail of game engines with bated breathe.

 

Unity is so under priced , and i 100% still stand by that.

I really, really hope Unity doesn't follow suit with this nonsense, stay classy Unity.

 

That's on purpose to make switching over from unity easier

 

Wow, knee jerk reaction is that UE4's business model is crazy tempting.
However, if Epic gets the support that I'm thinking they will, devs will subsidize development costs. As devs will have access to the source code, we shouldn't need to wait for bug fixes, as community debs will most likely have pulls that fix most common issues. Not waiting on Unity point releases to fix bugs that non-source devs don't have access to.

I never even looked at a tutorial for UE and just looked at the programming tutorial and it's pretty neat. It's definitely more programmer oriented it seems. With the source code download and compiling into the Editor to create a project.

Thing I will miss from Unity though:
C# - I wanted to forget C++ as soon as I could and now I find myself needing to brush up on it again.
.NET - LINQ, string, and an easy API will be much missed as STL was a poor substitute.
NGUI - I mean what is UE4's UI solution? Is there a similar substitute on the unreal marketplace besides adding the cost of scaleform?
Fast Recompiles - Say what you want about Unity's compile system, but after watching how long the "recompile" took to add a new property to a class (https://www.youtube.com/watch?v=Q3AvZmZEPyc). Means I can't shotgun my way to victory anymore.

 

Careful angrypenguin - changing an opinion after new facts were presented might break the internet.

 

You can also add a very important factor for people like me, here: Unity CAN still be licensed perpetually. Yes - I do want to license perpetually and not sort out royalties.
I know UE4 allows to download for 19.90$ and then cancel and keep. Which is a pretty strong thing to allow (I guess Crytek won't do this). Still. I like perpetual licensing and I really fear the day too many users start leasing their software.

I leave Crytek pretty much out of the equation for now because we don't know much about their final licensing model yet.

 

CRYENGINE: $9,90/month, no royalties, details to follow... http://www.cryengine.com/features/cross-platform

Of course you know this means war!

This is how indie games have changed the scene, and the most important thing to rememeber is that it's Unity, who was one of main powers in this process. And they must have known, starting with their policy years ago, that this day will come. In fact forcing competition to make the moves like that is their success.

So they are probably ready for that, or at least they should be. If they are shocked, that would only be their fault. But if they are as smart as they seem to be so far, then they already know how to respond. They problably know it for years, just waiting for the others to make the move.

They need to do something, that's for sure, the changes in the environment are just to big to ignore them. Doing a market analysis 4 days ago and today would give you 2 totally different outcomes, and when your market changes that much, you need to react.

 

I wouldn't say that. Unity drove the wide spread and ease of development. If there was no Unity, Unreal wouldn't now be attempting to compete in the market that Unity helped create.

Don't forget, Unity is still free. Completely free. You have to have a subscription to initially download UE4, you have to have a subscription to update, and to publish. And you have to give a percentage of your sales to Epic. Sure it is cheap, but all those things are free with Unity, and they don't take a cut. It is kind of the IAP of the game development.

Don't get me wrong, It's a pretty nifty thing overall, I subscribed, and waiting for the install to finish.

 

What a week this is turning out to be for indie developers.

To speculate on the possible impact, let's consider the user base from a technical and financial perspective over the short ( < 1 year ) and medium ( 1-5 years ) term.

It's unlikely teams working on existing projects using Pro or Free will do anything short term, as they are too committed and the training costs/time too great, so let's discount those for now. That leaves the vast customer segment of hobbyists on Unity Free. In the short term it is that customer segment most at risk of erosion. Only UT know what the average revenue per Unity Free customer is from the Asset Store. Will enough of an impact be felt to top line revenue to be of concern?

In the short term, new users now have much greater engine choice. Unity will continue to appeal to new users because of C#, JavaScript and Unity Free, with C++ presenting a barrier to entry onto other engines. Longer term these new users may have been up-sold to Unity Pro, so that is a lost sale opportunity which UT will be able to quantify in forecasts because they know their conversion rate.

In the medium term existing projects are completed, and teams start engine evaluation once again. Stickiness will be a factor, stemming from retraining costs, whether or not the developers wish to leverage any previous investment in tech they've built on previous projects, and the years of experience learning and working around the quirks. *Arguably* better graphics and performance, no Mono gc, and source code access are big pulls. But Unity is improving all the time in many areas.

No doubt UT have some head scratching to do, but I wouldn't expect a knee jerk reaction. If I was to anticipate a response, I'd be inclined to look to the pro subscription rate with possible packaging of the base product + add-ons, such as a Unity Pro Mobile Edition. I had previously evaluated the subscription, but was put off both by the cap on number of licenses that you can have on subscription ( no good for a team ), the higher cost over a version lifecycle, and to a much lesser degree that we wouldn't own the software should UT go under. I'd really like to see a renewal subscription for existing customers at a competitive price with no volume cap.

Will we be upgrading to Unity 5? Yes. Will we continue to use Unity after our current project is finished? That depends on what the next game is we want to make, how much we need to leverage the tech and knowledge we've built up with Unity, retraining costs, the level of commercial risk, and just how strong the pull is of other engines. To offer some perspective, we have a team of 9 ( 3 of whom use Unity ). The Unity license renewal costs for two years are negligible compared to two years of payroll. 5% of gross revenue would however be a sizeable cost, so the arguments for switching engines would need to be very compelling.

TL;DR - short term the battle is for the ocean of hobbyists who may buy from Asset Stores and who can be up-sold to. Medium term, we may see a migration of some smaller professional developers who's revenues are below the threshold in the OP, developers who may no longer be in agreement with UT's product strategy, and those with games that are much better suited to other engines ( FPS for example ). How important are headline titles to engine marketing? I have no idea.

Interesting times ahead, that's for sure.

UPDATE: The more I read into the Unity 5 feature list, the more I'm convinced we'll be sticking with Unity 5. Enlighten alone would be worth the upgrade cost.

 

Maybe UT will need to consider some price changes, but I think the important thing for them is to present what exactly makes Unity better. It should be clearer that their Enlighten implementation is very complete, for example. Some competing engines really don't have all that much built in, and I've read the state of GUIs in most engines is a sorry affair

 

i actually like this idea unity pro for development for free, but 2.5% - 5% of sales. and they be making money on unity asset store during development anyways of these people

at 2.5% even if they only sell $25k of games its $625 and if they sell 100K then its $2,500!

at 5% even if they only sell $25k of games its $1250 and if they sell 100K then its $5,000!

either way they making money on unity store and even more money the more they make. under above even w 2.5% at 100k its $2,500 which is more than a full license

they can have different % based on how many plugins u using. so for base pro it could be less and for all the platform plugin it will be more...

 

Obviously it's easier to explain when the people you're talking to are making money, and have gotten over free-lunch syndrome that's so endemic these days.

Without a vested interest - such as your livelihood - there's little to choose from between platform A or B other than which one looks prettier, or more fun to use, or has the most active community (obviously important if you're on the learning ladder).

As you know though, when the software is a means to an end, then things get more interesting, and complicated - ease of development, ability to scale, levels of support and so forth.

 

I just want to say that this calculation is valid if you have only 1 (!) seat... As per se, Unity wants your company to use Pro licenses on all seats if at least 1 seat is Pro... means that for my 4 man company the income limit would be $360.000... you actually add $90.000 per seat...

 

7. Unity has Linux export

 

Although for a small studio like yours UE4 is likely to still be better value (in purely financial terms) its not quite as clear cut. For starters the $90k usually covers about two years of Unity, so really you are looking at half of that yearly, so $45,000. Then after the first two years the cost drops by half again as you can get upgrades for Unity at half price or less.

However even ignoring the upgrade costs, you have to factor in that your team of 4 all want paying, how much do you pay them a year? On top of that you have to pay for studio space, utilities, taxes, to the point where I would have thought you could quite easily be needing to earn $180k gross p.a. on any game or games you sell in a year to stay solvent.

So while it sounds like a lot of money, I'm think many studios will end up discovering that the saving switching to UE4 is not as big as they thought. Obviously the break-even point is going to vary company to company, whether you need Pro add-ons, how much the company earns gross p.a., how much the company needs to earn annually to stay solvent etc.

On top of that once you start earning more than $180k p.a. you are technically going to start paying Epic more than you pay Unity.

 

IMO Unity should change the free version to 100% the features of pro with one limitation, you can only make money through Unity Cloud Ads, this limitation I suppose could just be enforced through looking through the code for any calls to places like IAd or the Google alternative.

 

Except Unity are only making money if the developer is making money and most hobby developers (who lets face it are the real targets of the UE4 initiative) do not sell $25k of games p.a. Sure there are dozens maybe hundreds of developers who do, but is that really enough to support Unity for the future?

Why should Unity give you everything for nothing on the 'promise' of paying them later, but only if you make money and only if you are genuine about opening your accounts to them.

Obviously it sounds like a great idea from this side of the fence, but put yourself in their shoes or imagine you set a business, would you really want to give away your work for free on the basis that you might get paid at a later date?

 

Interesting idea, might have some traction, but i'm unsure if the ad revenue would cover the loss of pro sales and i'm unsure if it could be protected enough to ensure people didn't hack the source to remove Unity Cloud Ads or add there own ones in.

 

The whole change in the market is about two things:

1. There is a choice now. Last week Unity was the only really affordable solution if you didn't want to give away 30% of your revenue. And now we have strong competition in the market, and the sole fact that we can make calculations what is a better solution is a very, very good thing for the market. Unity just got some real competition, and that will force them to improve their efforts, in this or the other way. Just like they forced Epic and Crytek, to improve on their strategies, by getting better and better.

2. Unity pricing looks "bad" on paper. Put $9,90/month next to $75/month and you will have a very hard time in convincing anybody that the $75 is the better deal. It is simply how people see things, how they see the prices. It's about marketing, the strategy, the image. It is Unity that needs to explain to the customers why it is 7 times more expensive than CRYENGINE, and this is not the position you want to be in as a company. It is not the product that got hit for the last few days, nor it's value, it's the marketing strategy that was attacked by the competition with a very smart moves.

All off the products are competitive, with their weak and strong points, but the whole point of the Unity marketing was: we are changing the world of game development by giving you the only really affordable and powerfull solution. But they finally changed the world, now they need to adapt to the environmet that they created.

 

No, I don't think it's a good idea to have gross percentages, because most people don't finish their games. It would mean that Unity basically gave away the engine for free to a bunch of amateurs without any sort of income guarantee at the end.

And a fixed price would actually be a selling-point. No need for follow-up accountancy. No need for Unity to monitor your books. You pay their one-time licensing fee when you have a game to sell. How is that not a positive? I think Unity could even move away from the 'per seat' pricing and increase the license fee. Like, make it $3000 plus $1200 per platform. Professionals would realise that it's a good deal, and Unity wouldn't lose as much money as I expect they will now.

But I do think the momentum will be lost if Unity keeps its 'up front', so that won't work any more. People will migrate to UE4, and a narrative will settle in that "everyone uses Unreal, don't bother with that Unity now". That's what needs to be prevented.

 

you mean like how UDK or cryengine let you use their engines for free for noncommercial uses for a couple of years now and when you wanted you could use it for commercial usage by having an commercial agreement with them? Or how other companies let amateurs use their software for free? Unity has been giving their engine for free.. they just have competition in the lower price points that are way more attractive. It is about marketshare and if you don't have users that care about your product, you shouldn't worry about monetizing. Unity still makes money of Unity free users off the assets store not to mention the tutorial and community it provides for them.

 

They'd have to match UE4's pricing model and conditions (or better) or they'd have to match Pixologics way of licensing of one-time purchase with upgrades for life.

If they dont, I'm simply dumping Unity once I wrap up the project I'm currently working on.

Its really a no brainer.

 

Unity may have been easier to use than UDK, but that definitely does not seem to be the case anymore with UE4, from the videos I've seen.

 

I've been using Unity for years and UE4 for about six months now (and CryEngine 3 for about a year), so I can say some things with confidence:

Hot reload / compile times
Unity's compile times are by far the fastest (a few seconds vs UE4's ~20-100 seconds, depending on how many intermediate code files UE4 needs to generate - this massively blows out when using GUI stuff because of how dependent it is on nested macros). UE4 fails hot reloads quite often, needing the editor to be rebooted. Unity Technologies have really nailed hot reloading and data persistence.

GUI
I'll get the chance to confirm tomorrow, but in its current state as I understand it, UE4's GUI - called Slate - is a crazy mostly undocumented mess understood only by the dev team (they used it to make the editor GUI) and partially understood by a few outside developers. Most have left it alone, and work straight with the HUD to draw textures and such. Epic doesn't have any plans to natively integrate Scaleform, and suggest purchasing your own license and integrating it yourself.

Visual scripting
Visual scripting is only available from the asset store as mentioned, and is pretty sketchy compared to UE4 / CE3. UE4's blueprint system blows CE3's flowgraph system out of the water in terms of usability, robustness (can do for / for-each loops, for example), and code dev time. Given an existing function you want to convert to a node, UE4 takes a few seconds, whereas CE3 takes several minutes at least.

Adding code / custom behaviours
Unity's still fastest for adding your own behaviours for objects. Once they're added, it's down to compile time / hot reload time, which Unity still wins out at. That said, performance during play for UE4 / CE3 code can run far more efficiently than Unity's, purely because of the AAA backend - but this only becomes relevant for very complex systems that leverage the efficiency of the backend systems, which most indie developers (to whom these new low prices are targeted) really won't need to massively leverage.

UE4's hard to get into at first, but once you get past the usual barrier of what functions to call and objects to use, you're golden. Workflow for adding code isn't as streamlined as Unity yet - you click File -> Add code to project, then get sent to Visual Studio with your new file generated, extended from some existing class you picked. Then you can tab back and forth between UE4's editor and the code, and hit Ctrl+Shift+Alt+P to recompile.

Important to note is what happens when your project can't find classes you've defined when loading a level. You can lose them by changing class names, or deleting files. When those classes are registered as missing during level load in editor, UE4 deletes derived objects from your level. Unity's behaviour in this regard is much better - leave an undefined script on any GameObjects referencing a missing script. There's no indication this behaviour will be changed.

Garbage collection
UE4's garbage collection can really catch you when you least expect it. Anything derived from a UObject (which is almost everything that's not a value type), if instantiated during play, will be GCed after the timeout defined in the relevant .ini file (for me, 60 seconds). The way to fix this is to mark any references to such objects with a UPROPERTY() macro in the line above their declaration if they're a class member. However, this solution works only for single objects, and arrays. Want a map of objects? Other kinds of collections? Nope, UPROPERTY() doesn't currently work for them (note: will confirm this for the current UE4 version tomorrow!). So you need to work around it by storing all elements simultaneously in a UPROPERTY()-ed array or you're going to get crashes from GCed objects going missing during code execution, despite any null reference checks you have in place.

Unity's garbage collection is far better for most indie-sized games in my opinion. You just need to be careful to handle instantiation, destruction, and runtime performance (needing object pooling, for example), which should be part of your considerations when creating any script / class / system anyway...

Addendum: (1st August 2014)
UPROPERTY() still doesn't support collections fancier than arrays. Support is suggested for some future release.​

Rendering and materials
CE3's material system is amazing aesthetically, but the workflow is needlessly tedious and you can't define your own shaders(!). Unity's material system is simpler, and it looks like the new physically based rendering system + built-in shaders will be even more powerful yet easy to use. UE4's material system just blows both of these out of the water, however. No question, there.

Importing 3D models
With their latest release, CE3 has really boned the workflow from 3DSMax to their engine. Their rc.exe - used to compile a 3DS model into a CE3-friendly .cgf model - works sporadically if at all. To my understanding (I'm not the modelling guy in any of my jobs), UE4 works with .fbx files exclusively. There are free converters out there such as the Autodesk FBX Converter, but something's usually lost in translation, such as materials or scaling. Unity works with a range of filetypes for meshes, which is definitely to its credit if you're in a scenario where .fbx isn't what you can produce / what you're given.

Addendum: (thanks, PolishRenegade!)
Unity converts ALL models into .fbx internally. So you might think you're using your .3ds native files but in reality, in behind the scenes you're just telling Unity to re-convert to .fbx at save. The advantage really lies in that you don't have to manually export it each time, can change per-file import options and Unity exposes the pre-process API for you to change, but that's it.​

2D games / applications
Of these three, Unity seems to have the market cornered still. I don't know of any plans for UE4 to include a specific 2D aspect to their engine to the extent Unity has. The closest Crytek comes is having announced a platform for mobile development some time ago, but it seems to have disappeared into the ether...

Addendum: (1st August 2014)
UE4 have a 2D engine roadmapped, but work on it is minimal, and it doesn't seem slated to compete well with Unity's 2D engine.​

Oculus Rift support (if you care)
Unity's got some really solid community dev time put into the Rift. CE3 had Rift support semi-integrated, but as far as I know it's still very sketchy and easily induces motion sickness compared to the implementation for Unity. I've heard UE4 has Rift support, but haven't seen anything concrete to make a good evaluation of its quality. UDK's Rift integration was pretty good up to the point it was cut off (presumably to be refocused on UE4 integration), so I wouldn't be surprised if UE4's integration was the same or better.

Addendum: (1st August 2014)
I've had some time to play with the Oculus Rift in UE4. From the editor, you can click Launch in Standalone Process, then hit the tilde key (~) to bring up the console, type stereo on, hit enter, and have the game enter Oculus Rift mode. I can't speak for how accessible it is via code without first looking into it, but I have noticed the Oculus SDK is integrated in both Unity and UE4 (no need to port the SDK as a plugin yourself for UE4, it's a built-in Editor plugin).
From the perspective of someone fresh to both engines just wanting to get Oculus support for walking around an environment, UE4's marginally simpler to implement. That said, the last character controller I saw for Unity's Oculus support was a bit hackish, relying on fixed delta timing for movement updates (literally using a hardcoded delta time of 1f/60f), which for me was more prone to inducing motion sickness than using Unity's built-in Time.deltaTime for movement.​

Documentation / support
CE3 - code documentation is sparse if at all existent, and assumes expert knowledge of underlying systems (eg using CE3-specific acronyms that aren't defined in source or online), support is practically non-existent even by the community (and many community posts are very wrong with info they give), online reference docs are often broken or out of date.
UE4 - Answers forum monitored (for now) by Epic employees, online forums, public wiki... ok wow that sounds familiar somehow.
Unity - ... oh, right Unity's already well into this Plus by far the most helpful, comprehensive, novice-friendly, up to date scripting reference I've seen anywhere. A ton of tutorials and learning resources, too.

If you're fresh to game engines and want to learn... Unity's still the strongest choice. UE4 might contend once its totally original community support model is taken up by passionate expert community members.

Addendum: (1st August 2014)
The Unreal Answers forum, though still monitored by many Epic employees, is seeing the same issues that Unity Answers has: an inundation of already-answered questions, write-my-code-for-me questions, poorly-explained-so-unanswerable questions, unanswered technical questions, and (more critically for UE4 than Unity) a growing number of "answered" questions that are now incorrect due to engine updates.
Support from expert community members for UE4 seems to be picking up, perhaps because of the engine's code being exposed for passionate programmers to dive into. That said, Unity's move for releasing regular bugfixes could prove good for the community... but an important point would be ensuring features don't break between builds. Perhaps a better move would be to integrate some of the more popular plugins as engine features - I've got an inkling that this is on the roadmap, can anyone confirm?​

Summary:

CE3 - $9 (is it a lock-in contract?), but as an indie dev I'd say you'll lose out in dev time and poor support / documentation
UE4 - $19/month + 5% of revenue w/ option to drop out and continue using latest version after 1st payment, pretty solid choice if you're not too concerned with GUI stuff or 2D stuff, still has many rough edges and deployment to non-PC platforms is seriously very sketchy... wait until full release in 2015 before putting all your eggs in this basket, I'd say
Unity - $75/month + addons for other platforms (locked in for a year), and amidst current competition really only best for simpler projects under tight time constraints (compile times + perhaps less model converting) / projects that want to work with 2D stuff



Edit (1st August 2014): deleted nothing, inserted a few addendums to keep info up to date.

 

The best potential new feature of Unity 5 is the "Unity Cloud"/cross promotion stuff, IMO.. at least for mobile game dev. For me, the biggest hurdle is not technology or even engine price, but marketing. It is impossible to compete with the large mobile publishers to get into the Top Charts right now, as they spend millions of dollars to get/stay in the charts.

With the new Unity Cloud and a bit of cooperation, we Unity indies can set up a system in which we can drive selected games into the Top Charts. It would take some coordination and a fair system (maybe we vote for what to promote each day), but it could be a real game changer if we put our minds to it

I really hope Unity delves further into this side of their business model (hope they are reading).

 

I just this morning got a quote for a Unity Pro license from a reseller here in Australia and was thinking "Wow, that prize is crazy! At some point I will need iOS and Android Pro and maybe WebGL later". Then I turn to the forums and see this. On the downside, I've been refreshing all relevant forum posts the whole day, not a very productive day, but I'm just too pumped! Hopefully something good for us customers will come out of this.

They could kill Unity free and go no royalty Pro at $30 - $40 per month, all platforms included. Maybe $500 - $600 up front? I know I'm just throwing random numbers there, I'm just suggesting based on what I would consider reasonable as a customer and long time Unity Free user. I believe users like myself would be on board with that, as we've already invested in learning the tools and buying assets. I don't know about new users, but that could keep most of the, at least mildly serious, Unity Free developers from jumping over to UE, not to mention it would convert us to paying customers. Otherwise, as others mentioned above, it's really a no brainer, Not only will Unity lose new potential customers, the current customer base will shrink and will probably mostly consist of kids with no money and huge dreams who cant afford $19 a month.

UE has a lot of catching up to do, in terms of tutorials, asset store, indie community, tools and language support, but they will get there. Unity is already there in all these things. If they act fast with more sane/aggressive pricing, they can stop the flooding. Unity 5 was just announced, and a day later all the headlines are about Crytek and UE, that alone says a lot.

We'd hate to say goodbye, I hope we won't have to...

 

I prefer Unity. I refuse to compare these things looking at the specs and license costs alone. I think I would rather just dump my mobile add-ons.

 

UE4 FAQ "If you access the source code, you'll see that there is early work on other platforms, including Oculus VR, HTML5 and Linux. You are welcome to extend and modify this code, and also build and ship games for these platforms."

 

Unreal vs. Unity, Russia vs. USA... I wonder what the world of tomorrow will look like after all the smoke clears.

 

Great post, thanks! Only 1 small addendum; Unity converts ALL models into .fbx internally. So you might think you're using your .3ds native files but in reality, in behind the scenes you're just telling Unity to re-convert to .fbx at save. The advantage really lies in that you don't have to manually export it each time, can change per-file import options and Unity exposes the pre-process API for you to change, but that's it.

 

Thanks for the correction! I'll amend my post.

 

I got to agree on the UI front with 19 Dowwa UE4, but I gather they'll improve it sooner or later.. UE4 is still a beta really.

 

If UT offered a subscription-based model at $30-40/month with no initial 12-month commitment, I'd actually consider subscribing in addition to the subscription I just bought for UE4. I'd love to evaluate Unity 5 Pro for a few months and compare it to UE4. I played around in UE4 last night for a couple of hours and the editor was quite intuitive and responsive. I must admit that the eye candy is also quite impressive. The lack of a UI framework is a bit off putting, but I haven't investigated how hard it would be to integrate Scaleform or to roll your own solution. I'm just getting into game development as a side-hobby but have significant programming experience, so I'm evaluating UE4 more from a workflow perspective (materials editor, animation, etc.)

I will say that I haven't really missed C++ since going to C#! As was mentioned in a previous post above, you really forget how C#'s features such as LINQ, generics, reflection, and the BCL in general make things so much easier (and fun.) I'm certainly not looking forward to using macro/#defines again or the STL lol.