What's wrong with Unity?

#21 I need to take another look at Unity.

For a laptop with a Core2Duo processor, 64 MB of dedicated VRAM , (256 MB total VRAM, partially shared) , 2 Gigs of Ram, Unity demos online have been slow, choppy, and VERY unimpressive. that "nature" demo flythough was just not impressive at all. Torque's environments, even ones using pixel shaders, run much faster even with pathfinding , physics, and lots of other things thrown in.

Torque uses texture compression; I'm not sure if Unity does or not but I was so unimpressed with the SPEED and PERFORMANCE of Unity that I have never taken it seriously as a serious alternative to Torque.

ThinkTanks, for example, is lightning fast in a browser on this same laptop.

#22

Quote:Aren't those the only two games that been shipped with Unity and didn't Fusionfall require the Unity staff to heavily customize the engine at the source code level to ship it? This is not something you can do, even with their high-end no-source $2000 license. You might say I'm exaggerating, but please, show me where I can buy a Unity game.

No. Hordes of Orcs, Tiki Magic Mini Golf, Big Bang Brain Games, Global Conflicts: Palestine and Global Conflicts Latin America, My Animal Centre for the Wii, a number of games on Shockwave.com including RC Laser Warrior, Extreme Cabbie, Downhill Bowling and MegaPixel, a boatload of widgets for OS X, plus more that I'm too lazy to look up at the moment. Oh, and a bunch of iPhone games.

As far as performance, the two Torque games that I've tried (that I know of) are Marble Blast and Minions of Mirth. The former runs well, but it's pretty basic. MoM is quite choppy when turning, so I can't say the performance is very good. It's not very hard to get good performance out of Unity if you know what you're doing; just about all professionally published Unity games and apps run smoothly. (This is from my experience on a 4-year-old machine.) Of course Unity uses texture compression, but generally other factors are a lot more important as far as speed goes.

The Pro license is $1500, not $2000. Given the few limitations with Unity, there's very little need for a source code license, for 95% of users. To say otherwise only shows a lack of understanding of what Unity can do, and to not take it seriously is a big mistake. And a source code license is available if you really want it (and have some bucks...).

As far as FusionFall goes, they switched from Gamebryo to Unity. They were using Unity for prototyping at first, but realized it was actually better to use it for production as well. It's only "heavily customized" in the sense that UT added some MMO features to the engine because of it, which would benefit anyone doing MMOs, not just one particular game.

#23

Quote:No. Hordes of Orcs, Tiki Magic Mini Golf, Big Bang Brain Games, Global Conflicts: Palestine and Global Conflicts Latin America, My Animal Centre for the Wii, a number of games on Shockwave.com including RC Laser Warrior, Extreme Cabbie, Downhill Bowling and MegaPixel, a boatload of widgets for OS X, plus more that I'm too lazy to look up at the moment. Oh, and a bunch of iPhone games.

I highly recommend that anyone considering Unity follow these links and play the demos of these games. They support my overall point. Unity is about where TGE was in 2005. Better tools and shaders, yes, but with no source, much poorer networking, and about 10x the price.

Quote:As far as performance, the two Torque games that I've tried (that I know of) are Marble Blast and Minions of Mirth. The former runs well, but it's pretty basic. MoM is quite choppy when turning, so I can't say the performance is very good. It's not very hard to get good performance out of Unity if you know what you're doing; just about all professionally published Unity games and apps run smoothly. (This is from my experience on a 4-year-old machine.) Of course Unity uses texture compression, but generally other factors are a lot more important as far as speed goes.

Good to know you're comparing games built on tech comparable to Unity. Marble Blast Gold and MoM were both based on 2005 TGE. MoM, even thought it's old and not so pretty by today's standards, supports hundreds of players on a single server. It's a real MMO with revenues to support a development staff...all built by two people on a shoestring.

Quote:The Pro license is $1500, not $2000.

I was trying to compare apples to apples. Since Unity uses a proprietary version control system that you have to pay for ($500), where as Torque INCLUDES source code (as any serious engine tech does) that can be managed using stardard free tools (like CVS, SVN, Git), the effective price to compare with Torque is $2000 unless you're working totally solo.

Quote:Given the few limitations with Unity, there's very little need for a source code license, for 95% of users. To say otherwise only shows a lack of understanding of what Unity can do, and to not take it seriously is a big mistake.

Maybe Unity is so magical that you don't seen source code, or maybe the source code isn't very pretty to look at, I don't know. I'm pretty skeptical that the engine is architected better than anything else out there (which it would have to be if source were really that dispensable). The results from the outside looking into the black box don't seem to support any magic claims.

Quote:And a source code license is available if you really want it (and have some bucks...).

LOL. Again, Unity must think they have some kind of extra special magic under the hood to close it down that tightly. I haven't seen any magic yet though.

Quote:As far as FusionFall goes, they switched from Gamebryo to Unity. They were using Unity for prototyping at first, but realized it was actually better to use it for production as well. It's only "heavily customized" in the sense that UT added some MMO features to the engine because of it, which would benefit anyone doing MMOs, not just one particular game.

Unless you've seen the deal terms on the Funcom license, this is a totally ridiculous comment to make. Only the Unity staff know what the did and didn't have to do behind the "binary wall" to make the engine comply with Funcom's feature spec. What is clear is that Funcom was not working with stock Unity, but actually getting development assistance at the source code level from Unity.

Hothead certainly didn't ask for, or need help from GG to make Penny Arcade Adventures. Licensing Torque means you're not bottlenecked by the Torque development staff. Licensing Unity means that Unity better support every feature you want in your game, or you'll have to wait for the Unity staff to get around to it. For a game developer, this sure seems like a broken model. There's a reason no other game engines are licensed this way.

#24

Quote:Unity is about where TGE was in 2005

What makes you say that? From what I have seen Unity is actually more comparable to the current TGEA 1.8 in terms of features. Networking is obviously the thing that stands out most in Torque, and I'm sure no would argue that. Could you outline the differing features for us?

If you are going to compare the $2000 price of Unity to Torque, make sure you also compare the features it comes with :).

Also I don't know why you even bother mentioning how 'pretty' Unity's source code is. That has nothing to do with anything, and from what I've heard it's probably a lot 'prettier' than TGEA source code. It is quite obvious that Unity has a completely different business model to Torque, not offering the source code has absolutely nothing to do with how functional or pretty their code is. I'll just file these comments under the 'grasping for straws' category shall I :P

As much as I love seeing the employees of each engine trying to put the other engines down - I wish you guys could keep it realistic and just use the facts and features of the engines. There is no need to start a cat fight over it.

Here are some examples of how you can do this:
- TGEA has better networking out of the box than Unity
- Unity has a better editor than TGEA
- Unity has a better (and working) PhysX implementation than Unity
- TGEA and Unity have both been proven as MMO capable
- Unity has a great pipeline, however this is only the case if you buy the Pro version with an asset server
- TGEA works very well with versioning control applications such as SVN
- TGEA has the source code meaning you can fix your own bugs and implement your own features (sure sometimes you may have to fix your own bugs)
- Unity doesn't have source code so you rely on them fixing bugs, granted they seem to be pretty good at keeping up with this in most cases.
- If Unity has all of the features you need, you WILL be able to make a game quicker than you will with TGEA.


How hard is it to do that sort of stuff? I'm sure you guys have done your homework and know what Unity does better and what TGEA does better. Instead of spewing out nonsense about 'magic', you could actually help people decide between engines.

I know, I know, it's all about marketing and you have to say your engine is better than all of the others, even if it isn't - but it really isn't very professional, most of us can see through it and it just lowers our respect for the respective parties.

To be quite honest TGEA has been the most difficult engine to get 'real' information about, you have to sort through all of the fluff and comments to find out the truth about the engine and it's out of the box - working - features. It is a real shame.

#25 reply to staples post:

well you said you have to "implement your own features" not really people have made lot of features already for torque and release it to the public ether for free or for a price. you might be able to build a game just buy taking other ideas that people have released already.

you also said you have to fix your own bugs but yet i see people posting help to fix a problem they made and i have always seen so far someone help them fix there problem so you don't always have to fix your own problems. gg is about its community and helping others when your building your game that's why i have said every time i go to these other engine sites i always turn back here as i know i going to need help with problems i might have. i not even going to look at least other engines anymore its not worth my time as i know i will turn around and had back to gg.

i really don't care what you decide as long as i know where i am staying i am fine with that. i just trying to help you decide so you don't go and spend a month deciding like me. they have made a post saying that they fixed allot of bugs today so i will be buying tgea soon.

#26

Quote:I was trying to compare apples to apples.

OK, in that sense $2000 makes sense.

I should also note that there are many self-published Unity games also available. I suspect at this point, there's more Unity content available than Torque content, despite the Mac-only status of the authoring tools (and that status won't last much longer).

For the vast majority of games people want to make and play, Unity provides far more than enough to do what you want. The API is very wide, and a certain amount of it is fairly low-level, so anything not provided out of the box can usually be built anyway with relatively minimal fuss in most cases; not that doing so is necessarily trivial, but it depends on your programming skill. (Some of Unity is actually built with Unity.) I sense a certain clinging to the notion that "if it's not hard to do, it can't be any good". For inveterate tinkerers who like to get their hands dirty with all the nuts and bolts, sure, in that case the source is desirable. But tools for game creation continue to get higher and higher level while still retaining most or all of the performance and functionality...some people didn't want to give up assembly language programming, either. (Personally, I don't miss it much, considering how much more complicated things are now than they were when I first learned it, although I do understand that mindset.)

Much of Torque's source code is fairly ugly, so I'm not sure how "pretty" enters into this. Whether to provide source or not is primarily a matter of company philosophy, not a competition to see whose source code is prettier or which engine is more "serious". And again, the source is available, so if you are in fact really serious and are making a medium-budget game, I gather the source license is still significantly cheaper than the $200K-and-up engines, so it's a non-issue in those cases.

#27
Brandon:

A lot of what you have said is speculation, sure I have found a few resources that look like they could be useful (albeit outdated), but the rest is just hearsay. In my opinion GG needs to put together a list of what resources are available, how old they are and if they work. It's all well and good if there are plenty of examples etc on the Private forums - but how am I supposed to know about these?

I agree that the community is pretty good and people will help you fix bugs, but even if this is the case, GG should be releasing a fix to the source with these bug fixes - something which they are not doing. If the bugs have been fixed, then why if I buy TGEA 1.8 now do I have to go through and fix them again. It doesn't matter who did the fix, it should be patched.

Btw I can guarantee the communities at both Unity and C4 are excellent, I have made a number of posts in all forums and get very quick responses. GG is pretty good for this as well. C4 is by far the quickest to get an 'official' response.


Don't worry I have already spent a month looking at engines, so these 3 were the ones that stood out too me. I have changed my mind a number of times, and have been trying really hard to find a reason to choose TGEA or wait for T3D.

At the end of the day, my biggest issue with TGEA is that it is so hard to find accurate information on what it can and can not do out of the box, and then what resources there are to fix the issues and help you out.


on a side note: It would be great if employees had "Employee" under their name instead of "Torque Owner", it makes it difficult to identify official posts. I know the forums are new, so maybe fix for this is already in the works.

#28 @Staples: There's having features as in "yes, our technology can do that" and then there's "yes, our technology can do that really well." Feature for feature, yes Unity and TGEA 1.8 look pretty close and I agree with most of the points you're making above list, but it does discount how well the features that both Torque and Unity have actually perform. Look at the Avert Fate demo. Look at the frame rate. Look at what's happening on the screen. Look at the visible distance. See if you can profile the texture memory, draw calls, etc. and look at the fidelity of the render. Then look at this (you can play it right now on InstantAction.com except that last shot):









All of the above is OLD stuff (like 2007 TGEA 1.0), but it runs great even on very low-end hardware. That's what I mean by performance. It's not just "we can do shaders" or "we have networking." It's how well you do what you can do.

And re: source code, you're right, Unity does have a different business model, but they don't present it that way. They attempt to say source code doesn't matter, or even more preposterous, that if your engine is built right, you don't need it at all. Have any friends who actually build games for a living? Try that out on any of them and see if it passes the laugh test.

#29

Quote:Have any friends who actually build games for a living? Try that out on any of them and see if it passes the laugh test.

Well, that would be me. I make my living 100% building stuff with Unity, and haven't seen the need for source code, nor do I feel like I've been held back...after 2+ years, there are still areas of the engine I haven't even touched. I don't see anything in those shots that make me say "Yeah, that wouldn't run very well if I made that in Unity". Avert Fate isn't the best example for how fast you can get stuff to run.

#30 just one last thing they are working on 1 bug fix for tgea 1.8 when will it be finished don't know as they got many other things they doing.

ya its hard to find resources right now i think on the private forums its easier though at lest from what i read. i have been saving resources when i find them as well. i also been saving sites people make with a list of resources not all of them just ones with stuff i like if i saved them all it would be hard to find anything when i go though them all. :P

#31

Quote:TGEA and Unity have both been proven as MMO capable

Unity is only MMO capable if you have the cash to drop for the source.

#32
I'm pretty sure you don't need the source, you can drop in SmartFox or NetDog etc with the Pro version, and the Indie version has RakNet and the .Net library if you want to roll your own.

Either way I don't think you need the source, but I haven't personally done it.

#33 These things can be argued 10 ways from sunday as the saying goes, and still nobody is going to agree. There's no point.

I am a fan of Unity so I say this with no ill intent. No matter how you argue it, no source code is a downside, and is limiting. This part can't be argued intelligently. Maybe your project doesn't hit the wall that's clearly there by having no source.. but it doesn't change the fact that the wall exists.

C4 is a very good engine, and one day will be great. If you have the time to wait (it will be more than 6 months), by all means go for it. Honestly, C4 is very tempting.
Unity is also a very good engine IMO. If you have the extra cash to get the license that allows for the stuff being compared, and the lack of source doesnt bother you, as well as the license structure, go for it. One day, I most likely will to.

But also like I have already said, TGEA (T3D) IMO gives the most bang for the buck. Agree or not, it doesn't matter. Go with what you think best meets your team's needs.

Peace

#34

Can anyone, then point me to ONE demo of something made in Unity that can compare with the speed and render quality of TGEA?

That "Tropical Paradise" demo is 2x as slow as comparable TGEA things in a browser. Full screen it simply does not compare to TGEA stuff I've seen. And "Tropical Paradise" does not have any physics other than maybe the camera collision sphere. And very few animated meshes. I didn't notice any stencil shadows either.

"Avert Fate" demo had the same issues. Slow and choppy frame rates on the highest graphics settings but also fairly slow and choppy on medium graphics setting.

And again, the overall quality of the experience, to me, looked like games built back in 2002/2003.

Torque demos, on the other hand, with the impressive sky and weather effects, shadows on all animated character meshes, faster physics (when grenades or collision debris are interacting with objects), etc.... just seems much more console like.

The machine I'm testing on is a medium spec machine: 256 MB VRAM (64 is dedicated) with 2 GIG system RAM, AMD Athlon X2 (dual core)

I just simply can't see where Unity compares to Torque in terms of speed and render quality. Torque seems MUCH MUCH more optimized in terms of graphic brilliance AND performance. But I'm open to hear anyone who can provide me with a Unity demo that would prove me wrong.

#35 Avert Fate ran fine for me.. didnt drop below 160fps.. but then my system isnt exactly low specs. It did glitch out causing the scope to always be in effect even when not using it. Some of the rendering look poor but that could be caused by several things.

#36 Yeah but my point is that I want someone to point me to ONE demo of something made in Unity that can compare with the speed and render quality of TGEA. One Demo.

#37 @Nmuta: As far a speed, keep in mind Unity is not optimized for any particular genre of game. It's emphasis is not raw speed (though it is a very fast general purpose engine) but more on flexibility, easy to use and being feature rich.

As far as render quality goes, these two threads on the Unity forums show some nice examples of what the engine is capable of visually (I believe the threads both have downloadable and/or in-browser playable demos as well):

forum.unity3d.com/viewtopic.php?t=17485

forum.unity3d.com/viewtopic.php?t=10486

My feeling is, both Torque and Unity are capable engines, both with pluses and minus, both with strengths and weakness ... use which ever suits you, and your project, best. :)

#38 Torque is clearly more advance than Unity,
check all these games made with Torque and prepare to be blown away :

www.vimeo.com/847225

plus with Torque you get the source code, so you can implement new feature when you need to, you don't have to wait for their staff to fix the bug for you.

#39 If you don't agree with this, you must live in a cave:

Quote:Given the few limitations with Unity, there's very little need for a source code license, for 95% of users. To say otherwise only shows a lack of understanding of what Unity can do, and to not take it seriously is a big mistake.

Just the fact that PhysX is implemented with networking is a HUGE difference. HHHHHUUUUGGGGEEE! Soft Shadows as well(pro version afaik) ? Well...WOOT!

ugh, there is no sense in arguing about aging tech vs. power tech.
- TGEA is a 54 buick with new rims and tinted windows. I want T3D.

#40 @eb:

Quote:If you don't agree with this, you must live in a cave:

Quote:
Given the few limitations with Unity, there's very little need for a source code license, for 95% of users. To say otherwise only shows a lack of understanding of what Unity can do, and to not take it seriously is a big mistake.

I don't live in a cave and I don't get this. 95% of Unity's users? Maybe, I don't know what their goals are. 95% of game developers...this is a serious lack of understanding about what it is to work in games. There's a reason why Unity's all alone in the world of binary development tools.

Quote:TGEA is a 54 buick with new rims and tinted windows.

Err...no. This doesn't even make sense. If you insist on the auto metaphor, it's probably more like a mid-90s Corvette with a faded paint job and a cd deck instead of a mp3 player. Easy to upgrade, but built on a high performance chassis with quality components, though maybe not the prettiest UI. But, as you mention, Torque 3D will be more like a '09 ZR1.