Official: How Can We Serve You Better?

It's unfortunate that Unity pawns it off on a Third Party, UnityVS has some real issues that they can not overcome because they are not integrated with Unity itself. Even just the simple fact that you have to add the unitypackage to every single Unity project that you wish to open in Unity after installing UnityVS is really S***ty and could and would be avoided if Unity themselves were making this...


I personally am really looking forward to Mono getting upgraded. I do a lot of Interop with Unity and my codebases end up being a horrible mess of basically:

Code (csharp):

1. #if UNITY_BUILD  (UNITY_EDITOR || !UNITY_64)
2.     // Some bullshit for the 32 bit version of Unity.
3. #elif UNITY_BUILD
4.     // More bullshit for the 64 bit version of Unity.
5. #else
6.     // The normal path for Microsoft .NET and... a modern Mono version.
7. #endif

The fact that UNITY isn't defined and I have to define UNITY_BUILD myself is also not great.


The fact that things like UNITY_64, UNITY_IPHONE, etc are defined while playing in the editor are a pretty large hindrance to myself as well as you can see above. When the 64bit editor gets here it will only get worse.

 

Thanks for the Reply Ralph and Unity team, I meant FIX MONO! Not Mono-develop *Sorry*, GC can be problematic. I know some Vita developers who have had a painful time due to this.

I'll also have to agree with Makeshiftwings slightly, some bugs have been going on for years. Some Unity could do nothing about, like Physx I don't think anyone can expect you to start fixing all third party SDK's. Others like the terrain system and Umbra / Mecanim, which for the most part I stopped using..

What is causing the major slow down in development generally GUI etc.? Reason I ask is, if we understand we won't continually ask about it.

Thanks Unity for taking the time, these little sessions really do make all the difference.

 

Autocomplete/Intellisense is really useful for me. It's the easiest way to find methods on Unity classes that aren't documented (which are useful, even though I know I'm probably not supposed to rely on them). I really like hitting the build button in Xamarin studio and getting inline errors, etc.

Find references, jump to definition are great. The xamarin studio refactor tools have been behaving well for the last few months, so I like that.

I only rarely use the debugger in Monodevelop. It seems much more awkward than using the debugger in XCode on our Objective-C cocoa apps. Depending on the day, maybe 1/4th of our Unity development is actually native obj-c in XCode, so we use xcode for debugging that, and that generally works pretty well. You can even see C# method names in the stack trace in XCode.

 

Oh and i forgot: I have question about Mecanim: there is any plan for make some humanoid-only feature available for generic type anim in 5.x cycle or later? (e.g. IK and retargetting anim(?) as far as i remember )

 

I'd like to post in contrast to DaFox and PhobicGunner on this, which I also kind of mentioned in my first post. I use UnityVS, and yes it is buggy but I like it much more than MonoDevelop. While in theory it would be nice for Unity to do everything themselves so that we never have to buy anything from third parties, in practice the more features Unity includes, the less time they can polish and fix each of those features. We have things like the Asset Store specifically so people can sell assets to step in where Unity doesn't want to. NGUI (and the other Asset Store GUI's) and UnityVS only came about because Unity was (sort of mostly) clear that they were not planning on doing anything about their existing GUI or VS support for a long time, so people saw the chance to make some money by filling that gap, and we got something that would work. I think this is important; if Unity is more communicative and open about what they are actually planning on doing and when it might be done, it would allow other people to see a business opportunity there and create the tools that we need. Case in point: UE4 said they don't plan on having any direct .NET support any time soon, and now there is a team working full-time on a .NET 4.5 solution for UE4, and they're further along than anyone has ever got with Unity. No one has bothered to try to update Unity's Mono because we have constantly been hearing that they're "working on it" but the timeline/actual solution is a secret, so no one has put much of an effort into finding their own solution because they might suddenly be put out of business if/when Unity announces whatever it is they keep hinting at. Bottom line: I prefer it when Unity honestly says "Sorry, go buy UnityVS if you want that; we're not doing it" than trying to be polite by saying that they "realize it's important" and vaguely alluding to potential future solutions.

 

I'm a co-founder at a mobile game studio (software engineer) who previously spent his career contracting for some of the biggest games companies in the industry.

Personally I'd love for there to be an IntelliJ plugin or some kind of collaboration with Jetbrains to facilitate this, and I feel a lot of other OS X (and no doubt Windows) developers feel the same way too:

http://plugins.jetbrains.com/wishlist/show?pr=&wid=193

It's been my tool of choice the past few years for every language that I professionally code in apart from C# (for Unity), and it's an incredible shame that there's no support - it's by far the best IDE that I've come across. I would love for you guys to be able to either work on or perhaps through some kind of partnership with Jetbrains to enable this to become reality, the latter of which could save time and effort on Unity's side to focus on more 'important' matters. Whilst Xamarin Studio is usable, not being able to have two windows of code next to each other (in a split screen manner) is a perfect example of the kind of basic functionality that it's sorely missing and a real hinderance on development - say when writing a unit test for example, not being able to have the test and the business logic sitting next to each other on screen is (for myself) a massive shame.

Thanks for this forum thread anyway guys -- this has managed to get me out of lurking

 

lol... replace. Visual Studio would be great. then you don't even have to worry about that crap anymore and trying to manage your own variant.

 

makeshiftwings:
I like it much better than MonoDevelop as well. And I agree with you allowing things like UnityVS and NGUI to exist is great. But I disagree with you on all the UE4 stuff.

The .NET integration in UE4 exists because UE4 is entirely open source. I'm not saying that doing the same thing with Unity is impossible, it's just not feasible. Even if Unity were not planning it themselves it's way too much work to be proding around while blind.

It's the same reason why UnityVS sucks. The UnityVS team is doing a great job with what they have, but ultimately their product will never be as good as if Unity themselves did it *or* if the UnityVS team themselves could integrate it more deeply with Unity.

A precise example from the UnityVS issues page:

An issue like that should never exist.

 

True, and I do think that's a big reason that Unity should consider shared source like UE4.

That was fixed, because the UnityVS team complained to Unity until Unity agreed to include the line number in the console metadata. UnityVS has a lot more traction when they whine about something than the rest of us because Unity relies on them for keeping their Visual Studio users happy. So yeah, there are definite benefits to something being directly implemented by Unity, but third parties can also have a strong influence once they become popular, so it's not always a dead end. I guess what I'm really saying is that the communication is important. "Yes, we are implementing direct Visual Studio support and it should be here in a year" is great, "No, we will not be implementing any direct Visual Studio support. We recommend buying UnityVS if you want that" is fine, but "We understand that our customers want a Visual Studio solution and your feedback is important to us" is the worst.

 

A few years back, I was working on a prototype, when my wife said to me, "You've been messing around with that thing for too long. Build something in 6 weeks," she said, and then in a softer voice, added, "You need this." Nine weeks later, I shipped my first solo product into the world. By night, I muddle my way through the world that is Indie. By day, I'm an nationally recognized game designer, author, and speaker on the topic of learning games. An Indie (3 years) and a professional (22 years).

* Unity Pricing - feels uncoordinated. As a pro, $4,500 is reasonable, and yet, as an Indie, paying for iOS/Android/Web add-ons really grates.

* Unity Subscription - is just wrong. Though the pro half would never be tethered this way, the Indie half was hoping the subscription could have been a godsend. Even with 200,000 customers, cash flow matters most. So half of me pre-purchased 5.x, and the other half still has 3.5.7.

* Unity Sales - With Martin Kharazi (sp) gone, so too is the personal touch. Expecting an invested human, I found an unresponsive 'enterprise' that showed little interest in me or my unique situation of owning TWO copies of Unity (day/night).

* 2D - GUI PARTICLES!!! Both halves love both 3D and 2D, and both halves want MORE 2D!

* Communication - Share a roadmap, be open, and do your best to meet a respectful timeline. If you think you're too busy, then consider that Tim F'ing Sweeney makes time to respond to individual users. (Recommended reading - 'The Speed of Trust')

* Open-Source - My old tech lead quit, rather than switch to a tech with closed source. Closed source no longer bothers me, though I think I'm in the minority.

* Features - Is the Asset store optional? As a pro and an Indie, I need core features that are only available as extras (ex GUI). Add that some assets are 'per-company' and some are 'per-editor', and it becomes a big, fat mess. Surely the Store sales hold a clue about which features could come next.

* Shuriken - More complete API implementation and support for 2D particle systems.

Here's to hoping that this new window of openness continues. Thank you for asking.
Gigi

 

Okay!

A.1) The biggest danger of doing business on the asset store is the number of 'dead' assets - that is, assets that have not been updated in some time, or ones with an inactive developer (that is to say, a developer who is either not supporting the asset, or not doing anything at all). Obviously you buy at your own risk in this scenario.

Something useful would be some sort of flag that informs the user that a particular asset/asset developer is inactive - while it wouldn't make sense to take the item down, this is information as buyers that we really need to know, for our own sake. This goes doubly for complex assets like code. I realize this is no small task, but this is perhaps one of the biggest problems on the Asset Store.

A.2) Another issue on the Asset Store is that there are some deceptive assets - probably the most famous one is the RatSpell MMO kit, which is known to be poorly documented, incomplete, and not work in even a partial capacity.

I'm sure the vetting team has their hands full with evaluating and critiquing all the new entries they receive on a daily basis, but I think that assets with naturally low reviews - like RatSpell - warrant communication with the developer to help them to improve their product. This may also help you to figure out if an asset developer is still active or not. This is beneficial for you as it improves the reputation of the Asset Store, and beneficial for us because A) we can actually trust Asset Store assets, and B) we can build games more efficiently with reiiable purchased assets.

 

As both a Mac and PC developer, I hope you'll continue to improve Monodevelop. It has major failings, such as the double-click-a-compile-error fail on Windows and the massive-memory-leaks-that-lead-to-crashes if you leave it open on a Mac. I cannot imagine why the best option is for game-devs to buy MS Visual Studio to write a Unity game...

Gigi

 

More to the second point, it was not uncommon to have David Helgason or Joachim answering questions on the forums in those days. Obviously those days are long gone, and it's probably not necessary to have devs answering simple questions, but they should engage the community periodically (vs never) with tips, hacks, etc. The blog used to do this, but even there, it's quite limited.

As top sub model: the 12 month lock-in is a deal breaker. It should just give you pro access, then if you cancel, it reverts to non-pro. Simple. Also, $225 for the (let's be honest about it) required licenses is a bit steep.

 

Another Asset Store quirk - I think Unity should run a quick test pass on its most popular Asset Store assets before releasing a Unity update to see if anything breaks, and consider fixing whatever bug was introduced that breaks those assets. Also, deprecate first, then remove, to allow Assets time to update. Some particularly annoying examples:

1) In Unity 4.2, you guys introduced a bug that caused all Tree Maker trees created in pre-4.2 to throw an "Invalid Mesh Serialization" warning and fail to display correctly. The only way to fix it is to manually click on every single tree and click the refresh tree button one by one. This affects all Michael O. assets and all the Big Environment packs, which make up most of the top 10 list of your environment section. Despite this, Unity offered no explanation of why it was happening, and marked the bug as Won't Fix for everyone who reported it.

2) In Unity 4.1 or 4.2, there was a bug that broke cloth physics, which basically broke every asset that included cloth and required they rebuild, but the rebuild wouldn't work in any earlier builds. Again, no real explanation of what the bug was, and it was marked as Won't Fix. Things like this require asset store sellers to constantly support multiple versions of their asset and the Asset Store has no legitimate way of handling that.

I feel like both of these could have been caught with even a cursory test pass. Making a change to tree serialization? Run a quick pass on your most popular tree asset. Changing cloth physics? See if it works on a popular asset with cloth. Does it break all the assets in the asset store? Consider backing out whatever it was that was changed, or at least put something in the patch notes about breaking changes for asset sellers.

 

+1.

Gigi

 

*grumble, grumble*

 

It seems that Mono will be replaced.
It's about time!

 

I'm CTO At Kurdify Technologies.. we have been using unity for 5 years now... what we struggle with is.

* That we don't know your road map... like back in 4.2 we created our own GUI.. as it seems its coming in 4.6 , we created Real-time GI , then we vanished it because its coming in 5.0 .

* I Really prefer a closed source software for these kind of works... we don't want to stick to 'nonofficial ' builds and fixes.

* More communication and frequent Hot-Fixes and Updates PLEASE.

* Unity feels like its Static. like the NavMesh... current G.I (talking about 4.x), Umbra's OC... most of them.

 

Except that you're not building from scratch, you have several working fairly robust systems to look at. I fully understand it takes time to iterate on a system but when a company as big as Unity takes a while to address such a needed thing as 2D and then the first release is not competitive with the current Asset Store offerings it makes me wonder how I can expect UT to stay competitive. There seems to be this Microsoft'ish "we'll get it right by version 3" type mentality going on with recent systems (Shuriken, Mechanim and the 2D system all spring to mind). There's nothing inherently wrong with this, but it certainly dampens my enthusiasm for new systems. For example, should I plan on using the new networking system for the multiplayer component of my current project? It may turn out to be amazing and wonderful and perfectly usable, but I'm not going to plan on that. I am a long time user of Unity, I have personally converted many, many people over to using it and sold plenty of Pro licenses for you, and if I am not very enthusiastic about new releases because I have lost confidence in UT's ability to deliver, that should tell you something about the hearts minds of (some of) your loyal users. Saying "yeah, but c'mon man it's version 1" is an excuse, and not really acceptable for a professional company of UT's stature.

 

When Tim Sweeney responds directly to a few clever inquiries every day, the community hears it, potential customers hear it, and the devs inside Epic hear it too. The message is loud and clear - customers come first.

Gigi

 

Re: Going open source

This single change would be enough to convince me to stay with Unity - and I can't be the only one. I'm not worried about the engine source (though others might want it), but even if we could open source the editor, we'd be able to make some great tools that enhance the workflow for other users.

Even something as simple as trying to extend the editor can be quite difficult and it'd be useful to be able to fork the editor source, make a couple of changes (such as exposing an internal property that controls which inspectors are available for the selected component) then send through a pull request. Whether or not it gets accepted will be up to UT, but it could certainly save a lot of time.

You'd also stick with your regular releases - source code should only be available to those who want it. Users may be worried about having multiple non-official versions of the source code, but this won't be the case at all - you'd stick with the official version unless you have an very good reason to use a non-official build.

I also love the CSG abilities of UE4 - and the keyboard shortcuts for snapping make it very easy to use. I know we've got ProBuilder in the asset store, but imagine how much more powerful and integrated it could be if they could have access to the source and suggest changes that would allow them to enhance their product even more.

 

Well when the news shows up I will take back all my grumbling, but until then I'm going to glower at my monitor so hard that the GC will crash trying to collect all my rage.

 

That's a great first post! The good thing about competition is it keeps you customer focused. Welcome!
Gigi

 

Reporting bias is an issue for sure. The busier I am, the less involved I am in the community.

My biggest gripe are the issues that fall between beta, bug, and feature request. For example, Joystick / Input support is terrible, and has been for two major releases. The beta list calls this 'out of scope', bug reports get terminated with "this is a feature request", and well, feature requests get kicked down the road in favor of marketable features. No one takes responsibility for the existing, broken systems.

My issue is that, as a user, this isn't a feature request. Joystick support is a feature Unity claims to have - it just is really painful (plug in four xbox controllers, and unplug one - or try to keep track of which controller belongs to which player). Who takes responsibility for implemented systems? New features are really beyond useless when the existing ones aren't adequate.

I know these are hard issues to solve, but we've all played games where this kind of thing is handled really gracefully - and when Unity puts their mind to issues, frequently really awesome stuff comes out.

As a user, I have no idea where to direct the bugs which extend into system changes, but really are critical issues in the runtime. I don't know who is responsible, who is tracking them, or what the known issues are. It's all opaque. Is Unity even aware how irritating the Input system is? I have no idea...

Even worse, the community frequently upvotes features which are trivially implemented by users. Granted, having expert help is awesome - but I'm more interested in Unity implementing the features we cannot trivially implement (e.g. requires native interaction with hardware across multiple platforms and device profiles) - abstracting that pain away is what makes Unity awesome.

I wish the existing systems had individual roadmaps, to acknowledge the fact that they are never complete. Software and hardware is constantly changing - and the existing systems need more attention.

 

I wasn't implying you just FIX MONO , I understand it's not that simple.

Well I'm quite intrigued to how you do it, I went down the wrapper / QT cross API / Xcode / PhyreEngine route and it was painful. But there is a lot of you.

 

Great to see the biggest crowd of Unity people in one tread, as somebody said: go out there and talk to the people. Nice.

I am a programer and little bit of artist form Serbia.

I used unity free but manged to buy pro this year.

It would be very nice not to have splash screen for the IOS and Android free plugins, but to use ( be obligated ) a little water mark " made with Unity "on the the splash screen. Something like UDK had before.

 

I've had that issue as well. Even something like.. "Hey this Unity method would benefit from another parameter" isn't a bug, so you can't report it via the bug reporter. But if it's an obscure editor function, it's got no chance on feedback.unity3d.com. No voting user is going to spend votes on something small when they could ask for Linux support or "more goats". Plus the contract I needed the change for will be over by the next major Unity release anyway, so I'm disinclined to spend the time writing a well-thought rationale for it.

 

I'm a student, who is hoping to make game development into a career. Right now, its a very serious hobby.

I have a few things conflicting here:
1) Unity is definitely worth the price, but
2) I would not have been able to buy Unity Pro without the student pricing. $1500 is just too much for people like me.
3) There are other products going for far less than Unity. I'd be lying if I didn't say I wasn't considering Unreal

I'm only interested in PC, at the moment.

I'm making two 3d games: A sci-fi FPS I'm almost done, and another early prototype game which would require about 16 different acronyms to summarize. Needless to say, the features required to it are similar to that of an FPS.

It would be really appreciated if devs stopped by more frequently than they do now, but I don't think it needs to be a daily thing.

Indifferent, but a subscription would ideally be cheaper than a perpetual license over the course of 2-3 years. I'd be fine with (reasonable) royalties as well.

I'm not an expert by any means, so I'd much rather the people who know what they're doing fix up the source code. That said, I've put off upgrading from 4.2 for my main project because of a glitch with occlusion culling that will be fixed in 4.5. Some sort of patching system would be great.

I'm largely a programmer, so any and all features that make art easier and graphics better would be greatly appreciated. Joints are also very difficult to use, especially when you're creating them at runtime.

 

Forget Mono, Visual Studio or Notepad. Visual Scripting is the future.
There is a huge crowd of potential users that dont know yet that they can build apps using an asset store plug-ins. They are now discovering U4, Even do playmaker seems a more elegant/ accessible solution than blueprints in their present iteration its not native its 'just' a 3party asset.

 

Allowing source access would be wonderful for bug fixing. I've had a variety of bugs that don't lend themselves to merely sending a project example and that leaves me in the lurch for how to proceed. An example would be this one.

It happens in different projects and is clearly machine specific but I really don't have any way of proceeding to look for the source of the bug. At least with source access I could try to dive into the relevant code to establish why it is happening.

 

Great thread.

@TylerPerry, arkon, L-Tyrosine, sebas77 - thanks for your comments on IAP, cloud, ads etc.

I lead the Unity Cloud team and we have been looking hard for many months now at the issues and challenges developers face and how we can help you find success once your game is released. Cloud services like analytics, cloud saving, IAP, virtual goods, social, ads and the like are all on the Unity radar. We'll have more detail to share in the months ahead.

You can always mail me at toddh at unity3d.com as well if you have something specific to ask in this area as well.

 

I somewhat doubt that.
Visual scripting is immensely useful for some things where you may not or should not want to involve a programmer (like level logic - push this button to open this door), but really shouldn't be used for scripting an entire game. Maybe a very simple game like Pong.

 

I sure hope it's not the future. I much prefer programming... if Unity had a built in visual editor as Unreal does, I'd probably make some use out of it for some tasks. But I think you just can't get the same power and control as you can with programming.

 

My biggest gripe is that I own Unity 4 Pro, IOS Pro and Android BASIC! I make about 10% on Android compared to IOS with the exact same games, yet for my $3000 for the 2 Pro licences plus 18+ months having to upgrade at $1500 a pop I still have a Unity Splash screen on Android! It looks so unprofessional

There really should only be one package for the Pro price of $1500 and it should include ALL platforms at pro level.

 

I can't disagree, but I see a couple of issues..

- Circumventing third party licensing.
- Abstraction and source modification.
- Upgrade paths.
- Knock on effects.

Unity will be very complicated and large, I'd say with all the platform support etc. it would be a bit scary for the best of us. But I'd rather have it than not..

 

imo on license side there should be three type of PRO - desktop (Mac linux win), console and mobile without any split on specific OS (like now we must pay alone for ios and android)

 

Please fix the Mesh API while you're busy updating the documentation for it. It's totally broken the way that it breaks everything up into individual verts/triangles that are not re-used. A basic cube has 24 vertices and 12 triangles and no edges at all! This prevents a lot of optimization like building occulusion volumes out of convex meshes without jumping through a lot of hoops to re-optimize and store the mesh separately.

 

Shouldn't a cube have 8 vertices, 12 triangles, and 12 edges? I'm confused as to why the Unity cube is so different than a geometric cube.

 

4verts = 8verts x 3 - each vertex is cloned for each triangle it is part of (so each face can have it's own set of normals etc). It's quite normal

 

Really there's nothing the Mesh API does wrong. I only wish it was more efficient (sometimes mesh updates can be fairly slow).

 

Squarely in the Hobbyist sector but owned Pro for 3.x, 4.x, and iOS Pro. I started with Free and quickly ran into roadblocks when I needed to use third party non native libraries as well as networking features. For my day job I've spent the last 20+ years doing everything from unix sysadmin work to network admin to network security(current), and when I'm really really lucky I even get to write some code to make stuff fall over.

Cheap sub definitely suits me, don't care about royalties either way. I would maintain it even while dabbling in other tools much like I do with Adobe creative cloud, my various VPS', etc. I'm not working on games to make money, it is purely for creative outlet and fun and anything I release would be free or relatively cheap once I considered it good enough. Before I ate the $1500, but with the recent competition I don't see a reason to continue down that path especially with android/webGL/Team license/whatever else comes out as a paid platform.

3D for Desktop. Dabbled in 2D for mobile and they are not the kinds of games I want to make.

Blogs are great, don't care too much about devs on forums as long as there is a robust question/answer repository such as Unity Answers. Updating that more frequently with dev responses would be more beneficial.

Not important to me, I prefer to always have the latest and greatest version of the software which subscription allows. Would not pay current Unity sub prices however.

Prefer source, have been involved with other open source projects and I believe in the model.

It may sound strange but the money pit that is the Asset Store. Even after I bought Pro I find myself using the asset store to replace the core builtin Pro functionality! Good example is shaders and postfx such as echoLogin's addons which are much more efficient than the included postfx and better shaders for mobile. Then I find myself buying terrain addons and terrain shaders, iOS plugins like Prime31, gui system(s), input managers such as cInput to work around input runtime mapping, 2d toolkits, networking libraries, pathfinding, etc. One day you sit down and realize you have this huge mess of addons whose main purpose is to make Unity engine core features work like they are advertised.

 

That's explained in the docs here: http://docs.unity3d.com/Documentation/Manual/AnatomyofaMesh.html

"To get crisp edges, it is necessary to double up vertices at each edge since both of the two adjacent triangles will need their own separate normals. For curved surfaces, vertices will usually be shared along edges..."

Cubes need crisp edges.

 

Unity expands all meshes, even ones you import, into individual triangles with duplicate verts and all, not just cubes.

 

Has someone requested mints on our pillows yet?