Unity 5 is coming and more!

Maybe! I love the fact you've put the paid version as no royalties. The only minor objection I have is the sub being 15 - value to people means different things.

 

This is for indies JUST getting started - they have literally nothing to call their own other than a copy of blender and maybe a student copy of photoshop if they are lucky.

That is why it is priced so low.

We also need to compete with Cryengine and Unreal but that is a whole other can of worms..

 

I'd say raise that to $30 or so. The reason UE is so low is likely because Epic still gets a great deal of income from their games, which Unity does not have.

 

As Unity are doing the upgrade for a year at $40.00 a month. They could keep that $40.00 going and add 3% revenue then make it available to everyone, also include Android and IOS in there. As for the rest do what HeadC says..

 

But then we have the problem of people defaulting to unreal or Cry as it is less expensive to get started. Just my 2 Cents.

EDIT: Lower it to thirty dollars USD and we got a deal

 

But what when they don't get along with C++ or whatever Cryengine is doing? Unity has made development really easy to pull off and has considerable resources and documentation available - I can't google anything on Cryengine's side.

 

You have a point Hippo. Lower the subscription to thirty dollars USD / Euro / Pounds Sterling a month and we got a deal for at least a year with unity tech.

Make it so!

 

Well, I for one am inclined to move on to UE4 if Unity should do away with Unity Free as I'd be involuntarily subsiding Pro users with the new subscription model. Really, I don't need a Lamborghini Testarossa on developing for devices with a 40 mph speed limit.

One should consider that $20 - $40 for a Unity Pro subscription and no Unity Free essentially ends Unity's reach into developing countries where that much money could be as much as a week to a month's salary when they have any at all and aren't subsistence farmers.

In my personal situation if Unity were to do such a thing I'd probably subscribe for a year and then move on, unless Unity Free were still available. That'd mean an end to my Asset Store spending which has been more then I even spent on Unity Basic licenses in a year.

More than Unity Pro it's the Asset Store that makes Unity money and doing away with Unity Free could and probably would cut significantly into that revenue stream.

 

I will admit that if Unity free does go away - Allot of thought should have been put into weighing the pros and cons as well as the potential after effects.

Just my 2 Cents atleast. I still think people should be able to use free if they do NOT make a payment or they just flat out cancel and their time is up.

 

I don't think they should get rid of free, I'm not sure if that would make sense?

 

I wish we could see more videos of Unity 5 being used. Would really like to see how great the scenes look in it besides that GDC trailer scifi corridors

 

Agreed - What I am thinking is everyone starts out with free and then they can upgrade to pro Sub or Pro Perpetual as their needs expand.

Just my 2 Cents at-least.

 

I just want to see Unity 5 released now

 

A rushed product is a bad product.

Give it time

 

Sandboxgods too busy on the Unreal forums slagging off my posts

But yeah, I share that sentiment.. I'm still hoping we can continue with Unity and avoid UE4..

@ HC88

Well UE4 is still in Beta, I had the import .FBX pipeline crash the editor a fair few times also the screenshot crashed it.. The LPV stuff still needs a lot of work.. But for all intents and purposes it's pretty cool.

As long as core features work, I'm fine with a few bugs..

 

LOL ShadowK! Ah I didn't realize that was you. Your name is a little different there

 

Lemme rephrase. I want to see an open beta (alpha?) of Unity 5

 

Ah Ok, I get you

Would like to see this as well

 

It's they're really releasing this year it's more like an open RC.

 

It's no problem I got my own in , then again that thread was out of control a bit. They have to realise that a lot of people moving to UE4 aren't hardcore UDK / UE / OpenGL / DX coders and a lot of them won't be used to C++ and they will find it daunting. Especially with people in the mobile market as a massive portion of them are probably Unity users. Source code is cool if you know what to do with it, which most people don't.. (Some guy saying he has the source code so he can port to any platform he wishes, sure I'd love to see him give it a shot). Personally I see a language as a tool to get the job done and don't have an issue with C++ doesn't mean others won't.

Then again there is Blueprint in it's favour.. I still think as a rapid workflow tool Unity is where it's at.. It's just a shame for bigger projects Unity lacks some of the key features and stability that UE4 has. Even with UE5, there is still some things missing, the particles system and AA is simply amazing in UE4. Plus something like matinee could really take the sting out of cutscenes..

I prefer the ease of terrain and grass in Unity, even if it is a bit long in the tooth..

 

*cough* update to the terrain engine? *cough*

 

Pretty clear to me it's still a few months away... starting to wonder why announcing so early then?

 

Probably because they didn't expected Epic or CryTek to release their engine for $19 and $10 bucks at this GDC. If they were the only big news for indie developers, they would've gotten lots and lots of pre-orders, and could keep the announcements at their own pace.

Now that we got news from Epic and CryTek (with Epic's engine being available right now), Unity 5 announcement seems early and lack meaty info as to what it actually is and when it is going to be released.

 

I don't think they will either. Unity have said repeatedly that free has been a massive benefit for customers and Unity themselves.

They have also said on multiple occasions that royalties are not a good idea. Collecting large values off a small number of customers is fine. Collecting small values off a large number of customers would be a nightmare.

The problem is it is a big step from free to Pro. Subscriptions don't a help as it works out more expensive in the long run and would be a minimum commitment of $3375 for me ($225 + 25% VAT per month, minimum 12 months).

I have said before that I would rather that each Pro feature would be available individually on the Asset Store. With some of them having a tick box in the editor for turning them on, they at least should be easy to compartmentalise. Maybe some interdependent features would have to be packaged together, but even that would make them more accessible.

 

Seems they've answered maintaining all the modules separately would be a lot of work for them.
I bought Unity Pro as soon as it came out in windows.
I'm older, have a good job, just another purchase for me and each version since then.
I've enjoyed the learning experience, found Unity actually useful (plus was able to hire a good programmer, expensive for my income, but glad I did).
Spent much more money on game relate programs that were nothing but a waste of time for me, due to the programming aspect.
All the time I was searching for a game engine I could actually use (I would have been happy to have a first person controller I could actually move about the environment), if I'd had a free version of Unity I would have been ECSTATIC, but guess many expect more now.
Sure, I'm no shining example of a game developer, still working on my first big game, maybe will, maybe not get finished,
but for me, it's a hobby. With all the zeal found here and many not able to afford Pro, guess the way to go, it like it's been said before, if your games make some money, buy Pro.
I know, should have just kept my mouth shut, but hey, I live here too.

 

Yeah. C++ is hard to learn, and even harder (impossible?) to master.
Even the pros can be prone to shooting themselves in the foot with C++
I'm willing to bet, for instance, that the recent "Heartbleed" bug was likely a result of using either C or C++, with the associated concept of how strings work in those languages.

The idea is that a string is actually just a pointer to the first character. Since this isn't enough data, you need another bit of data: how long the string is. This just tells you essentially how many bytes you can safely advance the pointer before you hit memory that doesn't belong to the string.

Now the way Heartbleed works is that, during a "Heartbeat" (basically, a ping), you send the server a payload you want it to repeat back to you, and a payload length. After receiving this Heartbeat, the software essentially "pastes" the string into memory somewhere and retrieves a pointer to the first character. Then, it advances the pointer for the exact number of bytes specified in the payload length. The problem, of course, is that you could give it a payload of three characters, but specify a significantly larger payload length, and OpenSSL will happily advance the pointer through memory that doesn't belong to the string. So the server ends up serving a partial RAM dump to the attacker.
I find it interesting that this problem would almost certainly not have been if OpenSSL had been developed in C#, since the concept of string pointer/string length simply does not exist (and, in fact, it's normally impossible to directly access memory in C#).

If that doesn't make C++ sound daunting, I don't know what would. I'm not saying they could not have fixed it (they could easily have used null-terminators to signify the end of the string, or even just check the length of the received payload), but that problems like this crop up in C++ a lot due to the very low-level nature of the language (Heartbleed is a very drastic example, a less drastic example is that it used to be possible in Twilight Princess to basically "stomp" on the program code loaded into memory by loading up a specially crafted save file with a sufficiently long character name)

 

Yes they have said that, although I'm not sure I agree.

Take Static Batching for example. That has a check box in the Editor. Enable if it has been purchased. There is probably more to it than that, but I doubt it would be that great a task to offer that without the other Pro features.

Same applies to Build Stripping, Splash Image on Mobile, and potentially more, if not all Pro features. Perhaps it would be worth testing with some of the easier to do options.

And if compatibility issues are stopping Unity offering those features on the Asset Store, should customers trust third party assets from the store?

 

Hey, if you can talk them into it, I have no problem with it, probably would save me some money!

I hope they don't use resources to maintain this set up they could be using to further the engine, though.

 

C++ isn't daunting for me at all, it's a chore though.

 

Well you've been at it longer than I have, It's not really surprising HC! You're like the silver age fox of games design and I thought I've been doing it too long (Feel old yet? HAHAH)..

Some people have used Unity for all there major projects and never touched C++..

@ Moonjump, the module thing couldn't work simply because people learning games might not really know what they need.

 

Yes, the Heartbleed bug was the result of using C for creating OpenSSL, because C lacks bounds checking. Basically what happened was a remote user could send a heartbeat request and the bug was in how that request got answered. The request contained a string to be used in the heartbeat and a number to let the server know how long the heartbeat string was. After receiving the heartbeat request, OpenSSL wrote the string to a memory location and then read it back and sent it to the user.

The problem was that OpenSSL trusted the number sent with the heartbeat instead of measuring the length of the string. Since C does not enforce any bounds checking on variables, this bug meant that somebody could send a one character string with a number telling OpenSSL that the heartbeat was 64KB long. Then OpenSSL would send use the pointer to the memory location where the string was temporarily stored to send back the heartbeat response of the length defined by the submitted number.

Without any bounds checking in C, OpenSSL was vulnerable to bleed what ever 64KB of data existed at that memory location. So an anonymous user could hammer a server with malformed heartbeat requests and randomly receive nearly 64KB of data each time. If a user did this enough times, the user would eventually gather a lot of data including the private SSL encryption keys, user login details, session data, and anything submitted through the forms (credit card details, etc).

The programmer made a mistake, but the language made it easy to make that mistake. The scary part of all of this is that people coding OpenSSL are extremely good at this stuff. If the OpenSSL team can make mistakes like this in C, then the rest of us are really screwed. I have a lot of experience with C, but I will openly admit that C is a dangerous choice for building any kind of security layer. The only thing I still use C for is writing firmware code for 8bit MCUs. That is it. Nothing else. The lack of bounds checking makes C too dangerous for large projects and/or projects where anonymous people connect to it.

 

Unity 4 was announced in June and released 5 months later.

--Eric

 

The Unity Tech guys work HARD, and right now nothing seems to be happening, but they always bring it out in due time.
Whether it's all you wanted or not, you will see continual progress in development of their engine.
Makes me tired just thinking about it, I would have given up long ago.

 

Learners who don't know what a feature does are not likely to buy it, just as they are not likely to buy Pro.

There are plenty of Unity customers who do understand what certain Pro features offer. You only have to look at the number of threads where people want one feature in free because it is not worth the Pro price just for that. Well let them buy one feature.

I think Unity are unlikely to try to compete directly on price with UE4. Offering something different that builds on a popular channel of theirs would be a much better way to compete, and would be much harder for the opposition to match.

 

Pretty much what I said

I believe there's actually a quote about this, by Bjorne Stroustrup

 

C and string (char) issues are well known so it's a real surprise to me that there wasn't bounds checking. It's as if someone just decided that would be cool to leave open. Any C programmer will tell you about buffer exploits, it's so common that I'm really surprised there isn't more to this story...

 

I wonder if OpenSSL isn't just using an underlying third party serialization library that just serializes strings like that (string+length) and they may not have given it any extra thought.

 

No...you can see the (fixed) code for yourself.

--Eric

 

This is wrong to blame it on C bounds checking. It's the programmer that designed the protocol not the C language. That said, I remember my days in C and the nerds that would use pointers and linked lists and such chicanery even when not needed but simply to be 'complex'. Ah, you got bit.

 

True, although there's no doubt that C made the security hole possible. It would have been impossible in many more modern languages.
That said, they do have their reasons for going with C. And they definitely should have been more careful.

 

No, before I even learned C I remember reading of a CS student that used Pascal to collect and paste together released memory and so reconstructed exams written by professors in word processors after the programs were exited. It's the same mistake.

 

No? No what? I didn't say C was the only language with this issue.

 

You are right that the programmer ultimately bears the responsibility. However, the language does play a role in this type of problem. C places too much responsibility on the programmer for managing memory, and that leads to problems like Heartbleed. And if the programmers working on OpenSSL cannot handle it, then nobody else can either.

 

If Unity is going to try to compete with UE4's subscription, then Unity needs to match the price. Epic charges $19/month. If Unity charges $30/month, then Unity will have a hard time selling those subscriptions. Unity really needs to start by adding a subscription option that is identical to UE4. That would reduce the risk enough that thousands of Unity Free users would jump on it. It would be a huge new revenue stream for Unity.

What I personally think Unity should do:
1) Unity Free (pretty much same as now, but possible with a few more Pro features)
2) Unity Pro Subscription - $19/month plus 5% royalties with option to quit anytime (no one year lock ins)
3) Unity Pro Perpetual - $1500 (and consider including mobile options into this base price)

This approach would easily lead to millions of dollars of additional revenue for Unity compared to what they currently have, because it would be easy to convince Unity Free users to try the $19/month subscription. Unity could shake all of the coins out of their pockets and make a huge pile of money.

 

The problem was the protocol if that's defined correctly there is no problem regardless of the language used.

 

Memory management in C# is ultimately handled by programmers are well so really C# and similar languages are creating illusions of false security.

 

False security?
Unless you're using code marked in "unsafe" blocks, you CANNOT randomly access areas of memory in C#. It's literally impossible. The aforementioned mistake (and yes, it was a mistake and I blame the OpenSSL programmers, not the language) would not have been possible to make in C# under normal circumstances.

 

It tend to agree but Unity has some time before UE4 can compete with the ease of Unity. I know Unity would get $20 - $30 every month from me and still get the asset store money. It all really depends on how many of those some odd 2 million Unity Free would subscribe and stay subscribed. I'm much more inclined to then, say, for example, a cable or satellite TV subscription. Unity is much better value than those.

 

Apples and oranges.

Unity gets you a bit of middleware you don't get with UE4. It doesn't matter if UE4 works with Enlighten; you don't bloody get Enlighten without paying for it. At least I haven't seen that company offer any such amazing deals yet

Enlighten, Umbra and FMOD are pretty nice ways to sweeten the deal. FMOD for indies is free for budgets under $100k, so at least that can be had with UE4. Let's see if Enlighten and Umbra end up with similar deals - UT would have to negotiate some nicer deals with each respective company then.

 

Some of this would obviously depend on how Unity's middleware contracts are structured. If Unity pays a flat rate for middleware, then Unity could easily launch a low cost subscription plan. If Unity pays a fee per Pro license for the middleware, then obviously Unity would need to add that into the cost of the subscription. Either way, Unity should try to match the terms and price of the UE4 subscription, since that is the obvious competitor.