uLink Free announced! Help us decide its features.

Dear Unity developers,

We at MuchDifferent are very close to releasing the feature packed uLink 1.3 (currently in
beta). You can already see the beta version in action in ShadowGun Deadzone. After extensive tests and observations, we are happy to report that it works flawlessly. With the official release, about a month from now, we will also change our licensing model so that studios are charged depending on how many of the uLink features they want. But the point of this thread is that we also want to provide a free version. With your help, we hope to define a feature set that is equal to Photon Unity Network (PUN) and provide it without restrictions, for no cost, to all Unity developers.

So, what we are hoping to do is ask you what uLink 1.3 features correspond to PUN’s features. Whatever is the consensus of this thread will eventually become uLink Free.

Please join or just read through our UnityPark Beta-mailing list for more detailed information about the upcoming features.

Thank you for your help!

//Christian Lonnholm
www.muchdifferent.com

 

i think the first part would be deciding how much or many concurrent users are allowed on the free version, most of them normally provide about 100ccu for testing. i would probably say around 150 or more just to beat the competition to start out.

 

This is excellent news, I've always been put off trying uLink through lack of a free version to develop with. The number of CCU's isn't real important to me as I would only see free being used to test features so a 20CCU would be fine for that and to test a few players, the main things are really the feature set to invest time in learning.

 

My question is how to prevent all the pirates from using up all your server resources? Do you have a way to authenticate legitimate customers?

 

I think you want to clarify whether you want to provide a hosted service (like Photon Cloud) or "just" the networking library (PUN). My understanding is you are talking about the later, but I think most people here seem to be thinking of the former.

Having used both libraries, for me, the big difference between the two is how you "customize" your server logic. Photon is all about having one big non-Unity server process with very optimized native network code, and uLink more about many smaller servers using the Unity framework, communicating with the uLink P2P protocol. It's thus quite difficult to "extract" the common features of both libs, since uLink without the Unity P2P part would be mostly useless.

For my current project, what I'd like for the free version is "just uLink" without all the others UnityPark Suite products (uLobby, uGameDB and such...) but the whole of uLink. If you want to put restrictions on that, maybe only allow N (2-3?) server processes using the same free license, and forbid P2P communication between processes with different licenses to prevent "cheating". I think it would translate rather well what the other libs call CCU.

If you are already thinking about some options and would like to discuss the things in private (like how much would I feel comfortable to pay for features X or Y), feel free to PM me here or e-mail me directly (you have my e-mail on my UnityPark Forum account, same username as here). Context : 2 person indie team, self-funded, developing small-scale MMO, with relevant industry experience. Disclaimer: for us, your current indie pricing is already quite a bargain, considering the quality of the lib.

 

First of all i should clear that MuchDifferent is providing a free uLink library and no hosting for free.
Secondly i want to provide my opinion on this.
1- I think feature pairity is not possible unless many features of uLink are removed. They should remove groups and scopes, custom data serialization and also peer to peer and selective secure RPCs and their bitstream and running unity on the server .

I think it's a good idea to limit number of users and disable some of the advanced features like P2P, OnPlayerApproval and maybe selectively securing RPCs and custom data serialization or some of these based on the community's opinions.

Obviously other parts like uZone, uGameDB, uLink and uPikko are not included.

Cheers to you guys as the nicest networking middleware for unity and one of the most awesome teams around.

 

You can already do it with the trial version which has no time limitation and only commercial releases aren't allowed in it.

 

Removing those things would basically make it worse than the built-in unity networking, though.

 

Oh when I checked before it was a 30day trial.. that's good to know I'll look into this again then Thanks

 

lol WOW is all I can say to this... What your saying is basically remove every thing uLink has to offer... WOW

 

Guys i'm not proposing removal of the features. i was just kidding and little and my point was that having the same amount of feature/value is not possible.
i think number of users should be limited to 100 or so
and P2P should be disabled because that's something advanced for MMOs and server to server communication.
Maybe making security like unity's built in network is a good idea because there should be a reason to buy it as well. They need money to move it forward and keep development alive.

The fact is that most of the money game engine providers make is not from these smaller sales but they are important as well.

 

I think it should stay as it is: trial license which in fact is not 30-day but unlimited time, and the commercial release of the game requires licence purchase. If you want to give a free version of ulink you shouldn't cut off any features and subsystems of UnityParkSuite like uGameDB or uLobby, because in many use cases standalone uLink is useless without those subsystems. Potential customers may come to the conclusion that your product is not the thing they want...
The CCU limit is also a good idea and it shouldn't be difficult to do, because there is a max users limitation in uLink.

So, for conclusion I would say: leave it as it is in current trial. Do not cut any features. Limit max CCUs.

Best regards and thanks for this wonderful system

 

@K0mbrain
When you are publishing a .NET assembly like uLink, piracy is really easy so not cutting features is really a big security risk for anyone in software business.
The free and paid structure of uLink seem something like unity's own at least to me. You have some features and will get others if u want it and pay for it.
In some situations i agree that having all systems is required for full evaluation but the documentation for all of them (other than full uPikko doc which is partially available) is fully available and i think some sort of trials will remain available for the paid version/versions as well.

 

Aside from Unity, I own very few free versions of software. Most free versions are crippled and make you suffer by lacking something essential to make your life easier or to make whatever you're doing "complete." This is why micro-transactions are so successful.

The issue to me always comes down to this... It's a business. It's about making money. When taking product develop, product maintenance, and support expenses into consideration, what business model makes the most sense? Do you want to charge more and have less customers, but those customers are serious developers who value their time and yours when it comes to support? - OR - Do you want to charge less, have more customers, but have to deal with people who don't value your time and have all the time in the world?

Free with limited CCU makes the most sense to me. If a developer is serious about getting uLink, then they just need a more flexible time table for testing and if uLink is what they want, they will buy it. If a developer will never have the money to buy uLink, then where is their game going with the free version? Is their game ever going to be popular enough to have more than say 50 CCU at one time? If not, then neither the developer or Much Different is going to be making any real money anyway, so giving a full functioning version with limited CCU isn't really giving anything away.

It's like giving someone who has a great idea for a cookie access to a fully staffed bakery for free - either everyone loves the cookie and it's going to make money, or everyone doesn't love the cookie, so no cookies need to produced - and thus the person just goes home. Is there an expense associated with giving someone access to the bakery? Of course! How do you make your money? By limiting the number of cookies the person can produce each day. Want to bake more cookies? All you have to do is pay $X.XX for each additional X number of cookies. Simple right? The person makes money, the bakery makes money, the customers love the cookies and everyone is happy. If only software were this simple. lol

 

Mmmmmmm cookies....... me likes cookies.

+1 for the thoughts though -- limited CCU fully functional would be my preferred options.

did someone say cookies? Mmmmmm

 

Hey Christian, I think it would be great to make uLink charged on a 'modular' basis. So you charge a certain amount for a uLink license, and additional license fees for each add-on module such as uLobby, uGameDb etc.. then perhaps offer a cheaper bundle price if the developer takes the entire suite.
Looking forward to the 1.3 release! We're about to release our game soon, now I must wonder if I should wait for 1.3

uLink has always been free to develop and test with. You only need to buy a license before you release your game with it.

Authentication doesn't necessarily need to be uLinks concern. There are many ways to write your own authentication service that provideds a token of some sort to an authenticated client, which can then connect to the game server. You can re-issue or verify this token at any time and boot anybody who is not authorized to be in a game session.

 

@MeltDown If your game required 1.3 features, just go and grab the beta. It's really stable and awesome and you can add features of it to your g ame or just gain some performance from new bitstreams and have many bugs fixed for you. Just keep in mind that both client and server should upgrade because there are incompatibilities between versions in terms of protocol and ...

 

Thanks Ashkan, I'll upgrade to 1.3 then.

 

Removing and limiting any part of uLink as it is now will make it much less interesting and annoying to develop.

CCU limit is the best way to go. uPikko and server / gamedb hosting is where you should make money from. There should be different versions of uPikko, basic/advanced.

But before doing all this, please update your scripts, tutorials/API's first. We spent so much time figuring basic stuff, its not even funny anymore. Those uLink smooth/strict scripts are sooo much more important, the whole networking thing is about those and you know its not really polished nor finished.

Its a sports car with bicycle wheels and a half finished instruction book how to change them. Sure it will drive and if you put enough time in it, it will be great. But what if..

 

CCU is probably the simplest way to limit uLink, and will be easiest for potential customers to understand. So I'd vote on this too.

Would also be great if you guys updated all your API's and documentation too, hopefully they'll be overhauled for 1.3 release.

You guys should also sign deals with server4you.com or leaseweb.com, from the research I've done, they are the most competitively priced and can deploy servers in EU and US in a variety of data centres. Those other guys you used originally or that you referred to me were quite expensive, so I found better value for money hosts.

 

I always thought that they're docs are the best among unity networking middlewares and are really good. Some of the more detailed stuff are not mentioned in Manual but other than that they seem good to me. However i can not comment on tutorials because i never used/needed them.
On SmoothCharacter script i should say that it works fine if you use auth servers specially.

 

Yeah sure it works, but it isnt there yet. If you open the scripts, you will see todo's comments everywhere. We spent a lot of time to "fix" stuff to our liking but thinks better scripts should be provided so we can use our time for the game instead.

 

They are, in my opinion they are actually the most comprehensive and well-written docs out there for a Unity 3rd party plugin.
They just aren't perfect though and some of the website pages are not kept up to date.

uLink is perfect, the docs and website should be too

 

Why uLink still doesn`t run on Linux?

 

Sure things can be improved and yes some pages are not updated.

Oh and Qualcomm's AR sdk docs are good as well.

 

It runs, uStream, uPikko and uZone are all written in Erlang and run on Linux fine and from Unity 4 you can export you game to Linux standalone and run uLobby and uLink servers on Linux as well.

You can run it on headless servers but need to install virtual graphics for running an X session (unity limitation which will be removed soon hopefully).

 

Utility scripts are great and having more of them is always better but basically it's your responsibility to spend your time in your game as you said and write your gameplay. The middleware should provide networking and MMO related features like persistent storage, load balancing, bandwidth optimization, networking layer and instance management and ...
It's something like unity's sample projects and standard assets to provide utility scripts.

But it's a personal opinion

 

Is uPikko available yet? When we contacted them about it we were told it wasn't even available yet. But it sounds like some of you already have access to it. We were told that the indie license wasn't available yet but they were interested in working out a Rev. Sharing deal with us but that uPikko was still being developed and wouldn't be released to the public for a while longer...

 

It's not available to us and i think no other indie team have it as well. The tech (pikko server) is surely usable because of their 1000 player FPS game but maybe it's not easy to use enough or packaged enough. I think it's not a kind of a product that they want to release to indies for a low price. After all no indies have the money for hose many servers for the dynamic big world that u want to manage with it. I think many want it to create something to atract investors but MuchDifferent might not be interested unless you provide some revenue share to them because you gained investment using their tech.

It's not a normal tech and is something really new so it's natural to me at least to be in this way maybe they have another word on it.

 

well we have the funding to afford a large amount of servers AND we already have $50,000 invested in Hardware. We just need the ability for the seamless world. We are on a tight deadline though so can't sit around waiting for stuff to be developed.

 

Add me on skype to talk on this if you are interested, maybe what u want is possible without uPikko and just with peer to peer and handovers and maybe the Pikko Server is ready atm.
You can talk to MuchDifferent again.

 

As a person who can from Photon to uLink, I must say all the I agree with this out of all the threads here. Eliminating features would be detrimental to software.

I'd say keep it as is!

 

Is there any update or thoughts from uLink on what model you are going for in the free version, and do you have a release date?

 

+1 for CCU limit. I tried Photon first for that reason alone. If they had better docs, better support, and their cloud servers had been less flaky, I might never have tried uLink.... (shudder)

As a solo startup, the ability to go commercial with limited users without too much capital outlay would be a huge help.

 

Their cloud can not run any kind of custom logic and is some sort of message relayer so even if it was at light speed maybe most projects could not go commercial with it

 

Don't worry, it does fine. We've got several commercial games live and we handle more than 10.000+ users playing at the same time for a title. Sure, not in a shared world but not all commercial games are seamless world mmo fps rpgs that need custom logic to handle more users.

 

Yes there are projects which can benifit from it. Chat applications and simple virtual worlds like that surely can use it but here we are talking more about games and usually games have logic on the server side. Even chess and simple card games need it if you are doing something real. So we are not talking about simless FPS worlds as a special case.

 

Custom "server side" logic in photon cloud would be handled by the master client in a room. The photon cloud is largely comparable to using uLink with the proxy server and allowing players to "host" their own games.

 

Exactly you are right, and who will do that? seriously! i mean who will do that for a real game? There is no single example of that. It's very different than allowing users to host their servers for casual multiplayer experiences or those who their main value is just having multiplayer and not competition and persistant stuff.
you should pay for badwidth and can not have a service reliable enough to do in app purchases and tournaments and ... for it to make money from. Let's put ping aside for now which is a real problem if your game is not turn based.

 

http://store.steampowered.com/app/207250/
http://store.steampowered.com/app/205550/

I'm not sure what you mean. It's nice because it works, all the time, not 30% of the time like with nat punchthrough, and it has no barrier to entry (eg: downloading server clients and hosting your own listen servers, which an incredibly small percentage of your playerbase is going to take the initiative to do).

 

I ment something from big studios. Yes you can claim that it's better than NAT in some games if you can pay the money for cloud usage and ping is less important but just that sort of thing.

 

You mean something like
F1 Online/Codemasters: https://www.f1onlinethegame.com/
Offensive Combat/u4ia: http://u4iagames.com/ (ex Activision Team)
Indestructible/Glu Mobile: http://www.glu.com/game/indestructible
...

But we LOVE Indies:
Uberstrike / CMUNE: http://uberstrike.cmune.com/
KOGAMA: http://www.kogama.com/
Critical Strike Portable/Studio OnMars: http://www.studioonmars.com/ (ex supercell)
...

all Photon powered.

 

I don't think that any of those big titles use photon cloud, are they?
I know your platform and titles well and watched familyguy online videos in unite and used your SDK before but we were talking on real usability of photon cloud in real environments in games which are realtime and require server side logic.
I was arguing against using master clients for games which their multiplayer part is a big one and asked for a big company doing that.

 

Is Much Different still in business? They haven't responded to this thread or updated their website this year.

 

They are in business, They just have released uLink 1.5.1 and the documentation is not up yet and licensing as well so you can go there and grab it and use it. It has good peer to peer and groups features added.

They worked on shadowgun deadzone and another title so used their tech and are improving it.

In their forums also they announced other quite interesting stuff which i don't know i'm allowed to talk about here or not, you can go to forum.muchdifferent.com to see them in questions.

So yes they are in business and if you go to blog.muchdifferent.com you will find some posts there as well.

There is no signs of stopping them, they have good contracts with swedish military if you look at their blog and they are quite faithful to their dev community however they might seem quite here. They are just much different but are available, highly available i mean.

 

Hello guys!

Sorry for the torturing wait We're actively working on this and discussing all aspects. I'm eager to show you the licensees we've come up with, thanks to your feedback. And thank you for your patience.


Sincerely,
Aidin
muchdifferent.com

 

@Ashkan_gc,

Thanks for the forum link. I couldn't find any Support / Community / Forum tab or link on their main website so I assumed they didn't have a public one.

@Aidin

Glad to get an update. I'm curious about uLink and looking forward to the release of a testing server.

 

They offer more products and services than just uLink... if you go to game technology -> unitypark suite you get the links to the manual and forum site in the sidebar.

 

Any news on the free version? I think uLink will become the go-to Unity networking solution if it becomes more accessible to indie devs. Looking forward to it.

 

A limited 100/150 concurrent users sounds perfect for evaluation.