Is there any alternertive to the webplayer plugin to provide unity 3D content to the web? is video streaming through cloud a possible route?
The unity webplayer requires the plugin to be installed to provide interactive 3d content on the browser through unity (or any other engine).
video streaming is a possibility but it will then just be a video, no interactivity, no realtime rendering, no nothing.
Not yet, but Google is working on Native Client, which allows native code to run in the browser. UT is working on taking advantage of that technology, so when it is released there would be no need to install a plugin for the Google Chrome browser.
And whilst this is great, assuming Unity would work properly inside Chrome for a change, it still needs users to update their browsers which according to the usergroups still can be quite a pain.
also native code isn’t a holy grail. Yes you don’t need to have a plugin, but the webplayer will just grow to include the engine on every single game unless google bundles it in which case it will really become a pain in the butt.
Also, Firefoxies already made clear that they don’t intend to support NaCl, IE will definitely not support it as they have enough plugin stuff making it complicated without direct access API and apple … who knows we will see.
means that for the other 90%+ of your users you still need the plugin
On the Video streaming , isn’t that onLive already doing that ? I mean with full input support, and also sounds like http://www.gaikai.com can provide a host service , but need to find some more detail .
No onlive isn’t doing that.
even onlive has a player.
The difference is just that your side of the player only captures input to send it to the cloud where the game runs and send back the output in an own codec etc which requires definitely an own plugin locally to read the video and show it
What this means is just that if you would go such a way you need to have several millions to build some rather massive clusters to do the rendering to send out only the videos instead of having the user render his own instance.
so for your situation no gain, just a death strike 
The rule is simple: If it is something thats not the OS or the browser in general that requires input handling beyond the offered things,it will always happen through a plugin cause you need code that captures and interprets input and interacts basing on it.
This holds for Unity, Flash, Shiva, Torque3D, OnLive, …
If you watch the unite video (around the 8 minute marker) they talk about bringing unity to googles NACL which means it wont need a plugin (well as long you are on the google browser – a catch 22), but it will also work on there “tablets and other devices” (well that sounds more interesting IMHO
Thanks for all the infos , I am well aware the google chrome runing native code/unity thing, whihc is still to be happening in the unknow future.
Yes, for onlive you do need to download a application. but Gaikai is claiming Gaikai is a cloud-based gaming technology that allows users to play major PC and console games like Call Of Duty or World Of Warcraft instantly, with one click. No download or install required. Thats what I am hopping for unity , Sounds like they (GaiKai) could support your own apllications to run using their tech, maybe pay them for hosting.
Then it won’t work with low latency if it does not use proprietary high end technology for codec etc ie meaningless for shooters and it will cost them a shitload of cash for the bandwidth of streaming the movie back. I wouldn’t hope for it to work reasonably.
WoW is one thing as latency is a low importance thing. But action games etc is unrealistic.
Thats the major hurdle that onlive has gotten above and even for them its very restricted in range etc as a latency beyond ~80ms makes action games unplayable / extremely frustrating
of course native client will also work with other browsers if it turns out to work as well as planned. “don’t intend to include it” means nothing while it is still work in progress and experimental. in worst case it won’t be built-in but a plugin for other browsers. even ie could have a native client plugin!
…and i guess chrome already has a much bigger install base as the unity plugin anyway.
don’t intend to include means that it won’t take place at a usable install base prior 2013 to 2014 and if you already think about this now, you will never have anything that will hit market
And unity has a plugin for any browser out there basically 
And chrome might have a larger install base but that doesn’t help you if only 0.01% of its users are using your game, still means that the vast majority would be using the regular webplayer plugin and that won’t change the next years. IE9 won’t have NaCl, the next IE won’t happen before 2012, so until then its clear as glass that you either have the webplayer plugin or commit suicide with a release.
yes, it probably still will take 2-3 years but you talked like it would never happen. and of course it makes sense to think about webgl, native client and such stuff already now. 
edit:
i think chrome / chrome frame aren’t a much bigger download than the unity webplayer or much harder to install.
WebGL: Sure if you like very slow execution speed, then yes. I think CopperLicht on the most current chrome gives you an idea on the performance.
Its a thread to flash 3D, but for unity, shiva, dx engine and alike its nothing they have to worry about
And 3-x years for game development means never as even MMOs project development only rarely take that long
Also, as mentioned, the Firefox devs already made clear that there is no plan on their side to support NaCl, which means the same for various others likely as they rely on the firefox plugin api to offer plugins at all.
Also NaCl is nothing that you can just integrate, the whole browser must be written for it basically as it is like a secure sandbox indepth browser and OS access layer.
Chrome is always auto updated to the newest version - which does sometimes pose a challenge for us to get the unity plugin to work, but should guarantee that once NaCl is available it will be on all Chrome installs.
Yes, this is a problem. Though, it may be partially solved by browser caches. Imagine the unity webplayer executable a binary which is downloaded from a single constant URL. Then, as long as it is used on a regular base, it remains in the browser cache, and won’t need to be downloaded on each use.
True, in the beginning this will only give us Chrome. Still, getting 10% extra plugin penetration is a good boost. If this turns out successful, it may spread to other browsers, either natively, or by use of a NaCl plugin. That’s something only time will tell.
Agreed, browser caches can help given NaCl allows using corresponding code from there.
Then its mainly a matter of explaining the user why it downloads from 2 distinct domains including potential security systems that trigger in in such a case as nacl stuff will surely end on the potential malware checklist rather fast.
Agreed, if it turns out successful it will find penetration. Would already be enough if apple integrates it into webkit actually.
Question is how long that will take. I don’t expect to happen it fast as we know how long google needs to solidify the stuff and prior to that happening it won’t even take off.
Also, they have to find ways to resolve the security risks, cause if it is such a can of worms and holes as Flash, why would anyone want to complicate the browser development for it.
the whole idea of native client is that it is inherently a lot more secure than normal browser plugins. that’s the reason the project is existing. running native code in the browser is nothing new. you can do it with activex and so on but then you get something like flash.